cartesia.ai

Meet Sonic-3: the best text-to-speech for voice agents|Learn moreMeet Sonic-3: the best text-to-speech for voice agents|Learn moreSonic-3: the best text-to-speech for voice agentsModelsnewAgentsSolutionsResourcesPricingContact salesSign inStart for FreeStart for FreeMeet sonic-3 for AgentsMeet sonic-3 for AgentsVoice AI like you’ve never heard beforeVoice AI like you’ve never heard beforeThe only streaming text-to-speech that laughs, emotes, and pulls you into the conversation.Try for freeContact SalesOh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.Oh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.ConciergeCustomer SupportCompanionGamingLogisticsEnglishPlayOh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.Oh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.ConciergeCustomer SupportCompanionGamingLogisticsPlayBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeContext-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Context-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Context-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Sonic responds faster than you can blinkHuman speedCompetitive advantageAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.Try for freeSonicBlink of an eyeHuman conversational response thresholdReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Sonic responds faster than you can blinkHuman speedCompetitive advantageAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.SonicBlink of an eyeHuman conversational response thresholdReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Sonic responds faster than you can blinkAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.Try for freeReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Powering agents across industries and personasBuild with Sonic-3Powers every kind of agent across industriesBuild with Sonic-3HealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreHealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreHealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreCurated voices for conversationFrom sidekicks to experts, our voice library spans every persona, helping you build expressive and engaging agents.Curated voices for conversationFrom sidekicks to experts, our voice library spans every persona, helping you build expressive and engaging agents.Instant & Professional Voice CloningInstantly create custom clones in 10 seconds—or generate Pro Voice Clones, fine-tuned and tailored to your business.Fluent and native, worldwideReach international markets with Sonic. It speaks 40+ languages covering 95% of the world, all with native voices. It even speaks 9 Indian languages—including exceptional Hindi.Explore 40+ LanguagesAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastDutchEnglish (British)FrenchGermanItalianPortuguese (European)Spanish (European)SwedishGreekFinnishDanishNorwegianAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastDutchEnglish (British)FrenchGermanItalianPortuguese (European)Spanish (European)SwedishGreekFinnishDanishNorwegianAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastPortuguese (Brazilian)Spanish (Latin American)English (American)English (Southern American)French (Canadian)Developer-first, enterprise-readyDeveloper-first, enterprise-readySonic is built for rapid prototyping and seamless integration. Developers trust it for secure, compliant, production-ready performance.Sonic is built for rapid prototyping and seamless integration. Developers trust it for secure, compliant, production-ready performance.Build with Sonic-3APIIntegrate Sonic directly into your product with simple, well-documented endpoints.SDKSpeed up development with pre-built SDKs in your favorite languages.PlaygroundExperiment with real voice interactions instantly in your browser. Test scripts, customize your voices, and hear the results in real time.Enterprise Grade Enterprise Grade SOC 2 Type IIHIPAAPCI Level 1Reliable uptimeMeet the teams we empowerDiscover success storiesRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersCookie SettingsLegalTerms of UsePrivacyAcceptable Use --- Meet Sonic-3: the best text-to-speech for voice agents|Learn moreMeet Sonic-3: the best text-to-speech for voice agents|Learn moreSonic-3: the best text-to-speech for voice agentsModelsnewAgentsSolutionsResourcesPricingContact salesSign inStart for FreeStart for FreeFlexible pricing for any scaleFlexible pricing for any scaleFlexible pricing for any scaleChoose a plan and use it how you like across our leading voice AI platform.Choose a plan and use it how you like across our leading voice AI platform.Calculate your pricingMonthlyYearlySAVE 20%All plans give you access toAll plans give you access toSonic,Sonic,InkInkandandLineLineFree$0/ monthGet introduced to ultra-low latency voice AI through core models and your own voice agentStart free20K credits for models$1 prepaid for agentsPersonal useDiscord supportSee all the featuresPro$4/ monthbilled yearlyUpgrade for instant voice cloning and to try voice AI in production for commercial useSelect Pro100K credits for models$5 prepaid for agentsInstant voice cloningCommercial UseSee all the featuresStartup$39/ monthbilled yearlyFor teams starting to use voice AI in production and need shared API keys, pro voice cloning, and multiple agentsSelect Startup1.25M credits for models$49 prepaid for agentsPro voice cloningOrganizationsSee all the featuresScale$239/ monthbilled yearlyFor businesses with large-scale use cases requiring high concurrencies and multiple agentsSelect Scale8M credits for models$299 prepaid for agentsPriority supportHigh concurrency limitsSee all the featuresEnterpriseContact usCustom supported models and agents with mission-critical guarantees for uptime, security, and complianceContact usCustom usage pricingCustom concurrencyEnterprise support via slackEnterprise-grade security & complianceSee all the featuresEvery plan comes with the best of CartesiaSonicSonic-3FLAGSHIP TTSWith a time-to-first-audio of 90ms, our flagship Sonic-3 text-to-speech model is designed for fluid, real-time experiences.Sonic-3 API accessVoice ChangerSonic-Turbo API accessInfillingVoice Library LanguagesDesign a VoiceTTS concurrent requestsSonic Text-to-SpeechVoice changerLocalizing a voiceInfillingPro Voice CloningInstant Voice CloningFreeProStartupScaleEnterprise23515Custom1 credit per character15 credits per second of audio225 credits as a one-time cost300 credits (one-time fee); 1 credit per character of infill text1M credits to train; 1.5 credits per character of PVC speech generatedNo cost to clone; 1 credit per character of IVC speech generatedSee usage ratesTTS concurrent requestsSonic Text-to-SpeechVoice changerLocalizing a voiceInfillingPro Voice CloningInstant Voice CloningFreeProStartupScaleEnterprise23515Custom1 credit per character15 credits per second of audio225 credits as a one-time cost300 credits (one-time fee); 1 credit per character of infill text1M credits to train; 1.5 credits per character of PVC speech generatedNo cost to clone; 1 credit per character of IVC speech generatedSee usage ratesTTS concurrent requestsSonic Text-to-SpeechVoice changerLocalizing a voiceInfillingPro Voice CloningInstant Voice CloningFreeProStartupScaleEnterprise23515Custom1 credit per character15 credits per second of audio225 credits as a one-time cost300 credits (one-time fee); 1 credit per character of infill text1M credits to train; 1.5 credits per character of PVC speech generatedNo cost to clone; 1 credit per character of IVC speech generatedSee usage ratesLineVoice agent development NEWLine is the modern voice agent development platform for building your first agent to your best agent—all in code.Line SDKText to AgentReasoning templatesTelephonyCall analyticsAccess to Sonic and InkBackground agentsGithub integrationCLIObservabilityNumber of agent slotsConcurrent callsCall durationTelephonyLLM usage during callsText-to-Agent CreationEvaluationsFreeProStartupScaleEnterprise13510Custom8122060Custom$0.06 per minute$0.014 per minuteFor a limited time only, LLM usage during text-to-agent calls are free of charge$0.05 per creationFree of charge for a limited time onlySee usage ratesNumber of agent slotsConcurrent callsCall durationTelephonyLLM usage during callsText-to-Agent CreationEvaluationsFreeProStartupScaleEnterprise13510Custom8122060Custom$0.06 per minute$0.014 per minuteFor a limited time only, LLM usage during text-to-agent calls are free of charge$0.05 per creationFree of charge for a limited time onlySee usage ratesNumber of agent slotsConcurrent callsCall durationTelephonyLLM usage during callsText-to-Agent CreationEvaluationsFreeProStartupScaleEnterprise13510Custom8122060Custom$0.06 per minute$0.014 per minuteFor a limited time only, LLM usage during text-to-agent calls are free of charge$0.05 per creationFree of charge for a limited time onlySee usage ratesInkInk-WhisperFASTEST STTAt just $0.13/hr on our Scale plan, Ink-Whisper is the most affordable—and fastest—streaming speech-to-text model.Ink-Whisper API accessMultilingual supportSTT concurrent requestsInk Speech-to-TextFreeProStartupScaleEnterprise8122060Custom1 credit per second of audioSee usage ratesSTT concurrent requestsInk Speech-to-TextFreeProStartupScaleEnterprise8122060Custom1 credit per second of audioSee usage ratesSTT concurrent requestsInk Speech-to-TextFreeProStartupScaleEnterprise8122060Custom1 credit per second of audioSee usage ratesTrust & Security for Enterprise PlanTrust & Security for Enterprise PlanContact usPriority Dedicated Support via SlackSingle Sign-On (SSO)PCI complianceCustom SLAsCustom Security ReviewHIPAA complianceEstimate your credits usage for STT and TTSproFreeProStartupScalesonicSonicInkcreditsCreditsCharactersDollarsminutesMinutesHoursFrequently asked questionsIf you question is not covered here, you can contact our team.Contact usFrequently asked questionsIf you question is not covered here, you can contact our team.Contact usDo TTS, STT, and Agent concurrency limits affect each other?How do model credits and voice agent rates work within each plan?How many Line voice agent minutes do I get per plan?How many credits do I need?How many credits do I need for Pro Voice Cloning?What happens to my rollover credits if I change my pricing tier?What happens if I upgrade to a higher subscription tier?What if I cancel or downgrade my tier in the middle of a payment period?When does my subscription renew?What happens if I use more model credits than I have in my account?What happens if I use more prepaid voice agent dollars than I have in my account?How and when are overages charged?How are break tags counted in billing?Is Cartesia SOC 2 Type II certified?Meet the teams we empowerDiscover success storiesRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersCookie SettingsLegalTerms of UsePrivacyAcceptable Use --- Meet Sonic-3: the best text-to-speech for voice agents|Learn moreMeet Sonic-3: the best text-to-speech for voice agents|Learn moreSonic-3: the best text-to-speech for voice agentsModelsnewAgentsSolutionsResourcesPricingContact salesSign inStart for FreeStart for FreeMeet sonic-3 for AgentsMeet sonic-3 for AgentsVoice AI like you’ve never heard beforeVoice AI like you’ve never heard beforeThe only streaming text-to-speech that laughs, emotes, and pulls you into the conversation.Try for freeContact SalesOh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.Oh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.ConciergeCustomer SupportCompanionGamingLogisticsEnglishPlayOh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.Oh wow, Valentine's Day snuck up on you, huh? [laughter] Don't worry—we'll get you a table, no problem! Let's make it special.ConciergeCustomer SupportCompanionGamingLogisticsPlayBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeBreakthrough naturalnessSo natural, it1laughs.It sounds palpably2excited.Sometimesdevastingly3sad.Itspeaksin42languages—like4Hindi.Andspeaks5justlikeyoumight.[laughter]Ohno![laughter]Thisisinsane!I’mdyingoverhere![laughter]Ijustcan't!Try for freeContext-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Context-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Context-savvy accuracy for the real-worldTry for free[01]CallupNASA,theFBI,andtheNSA.Then,um,tryUNESCO.Acronyms & InitialismsHandles acronyms and initialisms intelligently, reading them as words or spelling them out, depending on convention.Sonic responds faster than you can blinkHuman speedCompetitive advantageAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.Try for freeSonicBlink of an eyeHuman conversational response thresholdReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Sonic responds faster than you can blinkHuman speedCompetitive advantageAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.SonicBlink of an eyeHuman conversational response thresholdReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Sonic responds faster than you can blinkAt #1, Sonic sets the standard for ultra-low latency. It’s conversational AI that’s fast, fluid—and virtually human.Try for freeReal-time responsesSpeed designed for real-time interactions means conversations feel seamless, not laggy.Proven at scale, worldwideFrom San Francisco to Tokyo, Sonic leads in latency at P50 to P99 consistently and reliably.Performance budgetLow-latency from our text-to-speech creates affordances across the rest of your stack.Powering agents across industries and personasBuild with Sonic-3Powers every kind of agent across industriesBuild with Sonic-3HealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreHealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreHealthcareSimplify scheduling, clarify benefits, and enhance patient experiences with friendly, trustworthy voices.Learn moreCurated voices for conversationFrom sidekicks to experts, our voice library spans every persona, helping you build expressive and engaging agents.Curated voices for conversationFrom sidekicks to experts, our voice library spans every persona, helping you build expressive and engaging agents.Instant & Professional Voice CloningInstantly create custom clones in 10 seconds—or generate Pro Voice Clones, fine-tuned and tailored to your business.Fluent and native, worldwideReach international markets with Sonic. It speaks 40+ languages covering 95% of the world, all with native voices. It even speaks 9 Indian languages—including exceptional Hindi.Explore 40+ LanguagesAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastDutchEnglish (British)FrenchGermanItalianPortuguese (European)Spanish (European)SwedishGreekFinnishDanishNorwegianAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastDutchEnglish (British)FrenchGermanItalianPortuguese (European)Spanish (European)SwedishGreekFinnishDanishNorwegianAmericasWestern EuropeEastern EuropeAsia PacificIndiaMiddle EastPortuguese (Brazilian)Spanish (Latin American)English (American)English (Southern American)French (Canadian)Developer-first, enterprise-readyDeveloper-first, enterprise-readySonic is built for rapid prototyping and seamless integration. Developers trust it for secure, compliant, production-ready performance.Sonic is built for rapid prototyping and seamless integration. Developers trust it for secure, compliant, production-ready performance.Build with Sonic-3APIIntegrate Sonic directly into your product with simple, well-documented endpoints.SDKSpeed up development with pre-built SDKs in your favorite languages.PlaygroundExperiment with real voice interactions instantly in your browser. Test scripts, customize your voices, and hear the results in real time.Enterprise Grade Enterprise Grade SOC 2 Type IIHIPAAPCI Level 1Reliable uptimeMeet the teams we empowerDiscover success storiesRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyRavi KrishnamurthyVP Product“Cartesia’s state-space models bring enterprise-grade speed and quality to our AI Voice Agents.”Read the full storyBob SummersCEO“Sonic is the only product in existence with model latency of less than 100 ms, outperforming its next best alternative by a factor of four.” Read the full storySami ShalabiFounder & CTO“Cartesia’s Line platform provides us with the essential foundation for building voice agents for modern enterprise environments: speed, reliability, and truly natural voice interactions.” Read the full storyKwindla HultmanCEO“Cartesia Sonic is the best voice model today for real-time multimodal use cases.” Read the full storySpencer ChanHead of Poe Product“With Cartesia's Sonic model, users can interact with a wide range of high-quality, human-like voices in multiple languages, enhancing their experience on our platform.” Read the full storyVipul Ved PrakashCEOCartesia is leading the charge of building efficient, multimodal models from first principles, starting with their Sonic TTS model. Read the full storyHassaan RazaCEOCartesia’s Sonic model is a game-changer [...] Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations.Read the full storyReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersCookie SettingsLegalTerms of UsePrivacyAcceptable Use --- Meet Sonic-3: the best text-to-speech for voice agents|Learn moreMeet Sonic-3: the best text-to-speech for voice agents|Learn moreSonic-3: the best text-to-speech for voice agentsModelsnewAgentsSolutionsResourcesPricingContact salesSign inStart for FreeStart for FreeThe modern voice agent development platformThe modern voice agent development platformLine is a code-first ecosystem to get from zero—to your first agent—to your best agent, in record time.Build nowRead the docsWays to get startedStart from anywhere. We’ll meet you thereBring your own code, start with a template, or even a prompt. Line meets you where you are and gets you live fast. All roads lead to a code file that’s written in the Line SDK, so you can use the ecosystem to iterate.Get startedBring-your-own reasoningBring the full complexity of your existing production-grade reasoning system or start anew.TemplatesJumpstart a new voice agent with ready-made templates.Text-to-AgentGet to a v0 voice agent—with its code—through just a sentence.Bring-your-own reasoningBring the full complexity of your existing production-grade reasoning system or start anew.Using Line’s public SDK, you can quickly turn your knowledgeable text chatbot into a voice agent or start from a clean, easy foundation. Connect to any LLM system and tool call backends you want.TemplatesJumpstart a new voice agent with ready-made templates.Clone one of our templates, built on top of our SDK, and customize it to your use case, starting with outbound calling and inbound help desk.Text-to-AgentGet to a v0 voice agent—with its code—through just a sentence.Generate a voice agent with our client. Write your own text prompt to spin up an agent that you can extend via the Line SDK.Line overviewEnd-to-end—it’s one continuous Line of voice innovationLine lets you build, deploy, and test enterprise-grade voice agents on a consolidated stack that’s fully owned, fully optimized.Multi-prompt configurationMove beyond simple prompt configurations, so you get the smartest agents on the market.Tool calling for knowledge and actionGive your agent powers, including RAG capabilities for access to live knowledge that keeps your agent relevant and up-to-date.Background agentsCoordinate background agents that can listen, analyze, and summarize your conversations—all while writing to background systems in parallel.BuildRobust SDKOur SDK is everything you need to build the most powerful, intelligent agents with taste. Get advanced reasoning, parallel background tasks, and real-time system actions—all with ultra-low latency.Multi-prompt configurationMove beyond simple prompt configurations, so you get the smartest agents on the market.Tool calling for knowledge and actionGive your agent powers, including RAG capabilities for access to live knowledge that keeps your agent relevant and up-to-date.Background agentsCoordinate background agents that can listen, analyze, and summarize your conversations—all while writing to background systems in parallel.BuildRobust SDKOur SDK is everything you need to build the most powerful, intelligent agents with taste. Get advanced reasoning, parallel background tasks, and real-time system actions—all with ultra-low latency.Sonic text-to-speechWith time-to-first-audio under 90ms, Sonic reliably delivers realistic, fluid conversation that flows naturally. Sonic is the fastest model available on the market, affording ultra-low latency performance all around. Ink speech-to-textOf all streaming STT models, Ink has the lowest time-to-complete-transcript and is tested against real-world conditions including noisy backgrounds to deliver accurate transcriptions.ModelsTop notch in-house modelsOur fully-owned voice stack is underpinned by our leading Ink STT and Sonic TTS models plus deployment so you can fine-tune, co-locate, and optimize for max performance.Sonic text-to-speechWith time-to-first-audio under 90ms, Sonic reliably delivers realistic, fluid conversation that flows naturally. Sonic is the fastest model available on the market, affording ultra-low latency performance all around. Ink speech-to-textOf all streaming STT models, Ink has the lowest time-to-complete-transcript and is tested against real-world conditions including noisy backgrounds to deliver accurate transcriptions.ModelsTop notch in-house modelsOur fully-owned voice stack is underpinned by our leading Ink STT and Sonic TTS models plus deployment so you can fine-tune, co-locate, and optimize for max performance.Github integrationGet one-click deployment and scaling with Github integration in our reliable infrastructure.CLIRemain in your console when developing and run commands for a streamlined workflow.ObservabilityView logs from your code for every call placed on Line.Fast deploymentTalk to your agent in under 30 seconds.DeployFast deploymentThe best agents are built via rapid, iterative development. Code → Deploy → Test. Repeat. Line is designed for rapid development loops.Github integrationGet one-click deployment and scaling with Github integration in our reliable infrastructure.CLIRemain in your console when developing and run commands for a streamlined workflow.ObservabilityView logs from your code for every call placed on Line.Fast deploymentTalk to your agent in under 30 seconds.DeployFast deploymentThe best agents are built via rapid, iterative development. Code → Deploy → Test. Repeat. Line is designed for rapid development loops.System metricsReview key metrics like call success and time-to-first-audio.Call analyticsUse an LLM-as-a-judge to evaluate every Line call.Live testingInstantly call your agent on the phone or over the web for testing.TestBuilt-in evaluationsHow do you know your voice agent is any good? We added a framework for live testing, system metrics, and custom call analytics. System metricsReview key metrics like call success and time-to-first-audio.Call analyticsUse an LLM-as-a-judge to evaluate every Line call.Live testingInstantly call your agent on the phone or over the web for testing.TestBuilt-in evaluationsHow do you know your voice agent is any good? We added a framework for live testing, system metrics, and custom call analytics. Trust and securityEnterprise-ready with production-grade agents at scaleDeploy with trustReliable and always onGet dependable uptime and priority support with custom SLAs.Compliant and secureSOC-2 Type 2, HIPAA, PCI Level 1 compliant, with support for SSO.SOC 2 Type IIHIPAAPCI Level 1Flexible and privateDeploy flexibly to meet compliance, residency, and security needs:Secure APIManaged in-VPC (on-prem)For every industryBuild the most intelligent voice agents, custom-fit to your industryHealthcareImprove patient management, provide faster answers on benefits eligibility, and maintain HIPAA compliance. Build Line voice agents toSimplify patient requestsStreamline intakeAnswer questions 24/7 on billing, claims, and benefits eligibilityLearn moreFinanceReduce call center costs, modernize IVR systems, and maintain PCI compliance.Build Line voice agents toHelp with wealth managementAutomate FNOL and reduce paperworkHandle inquiries 24/7 for banking, credit card, and loan accountsLearn moreHospitalityGenerate more revenue, simplify operations, and improve guest satisfaction.Build Line voice agents toAutomate bookingStreamline check-inImprove customer service with AI-powered turndown service and wakeup callsLearn moreEmpower your business with voice AIGet startedReady to build?Ready to build?Generate intelligent, fluid conversational agents for just $0.06/minute on any plan.Read the docsContact usReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersLegalTerms of ServicePrivacyAcceptable UseCookie SettingsReal-time, multimodal intelligence for every device.ModelsSonicInkAgentsSolutionsCustomer serviceLocalizationRecruitingSalesFinanceHealthcareGamingHospitalityRegionsAsia pacificBrazilChinaIndiaJapanKoreaLatin AmericaMiddle EastNorth AmericaWestern EuropeEastern EuropeResourcesBlogCustomersDocsEventsPricingResearchSupportCompanyAboutCareersCookie SettingsLegalTerms of UsePrivacyAcceptable Use