Skip to content
Cartesia

Cartesia text-to-speech for Bolna AI voice agents

Cartesia is the text-to-speech provider that turns your Bolna agent's replies into speech using its Sonic models, which cover 42 languages and respond fast enough for live phone calls. Pick it when time-to-first-audio matters most: Bolna's provider guide lists Cartesia as the lowest-latency option for English.

How Cartesia fits in the stack.

  • Best for

    Turning your agent's replies into natural speech.

  • Use this layer when

    You want to choose the voice, language, accent or latency your callers hear.

  • Connects to

    The LLM's replies upstream, and the phone call downstream.

The synthesizer is the
voice layer.

Cartesia turns the agent's replies into speech, so callers hear the voice you picked.

  1. Telephony

    Phone network

    • Exotel
    • Plivo
    • Twilio
  2. STT

    Listener

    • Deepgram
    • Sarvam AI (STT)
    • Soniox
  3. LLM

    Reasoning

    • OpenAI
    • Anthropic
    • Google Gemini
  4. TTS · You are here

    Voice

    • Cartesia
  5. Tools

    Actions

    • Cal.com
    • n8n
    • AiSensy

How do I set up Cartesia with Bolna?

Optional: connect your own Cartesia API key at platform.bolna.ai/auth/cartesia. You can also use Bolna's bundled access.

Which Cartesia models and languages does Bolna support?

ModelWhat it isLanguages
sonic-3Stable Sonic 3 release42, including Hindi
sonic-3.5Stable Sonic 3.5 release, recommended for production42, including Hindi
sonic-3.6Sonic 3.6, the newest Sonic model42, including Hindi

For every setting, see the Cartesia setup guide in the Bolna docs.

All three models work with every Cartesia voice available in Bolna. When you create an agent through the API, language is required in the Cartesia synthesizer config.

What can you build with Cartesia and Bolna?

  • Fast-paced support lines

    Short, quick replies keep busy inbound queues moving without long pauses between turns.

  • English and Hindi agents in one flow

    Use Cartesia for English and another provider per language in a multilingual agent. The Bolna docs give the example of Deepgram transcription with Cartesia for English.

  • Branded custom voices

    Clone a voice from a clean 1 to 2 minute recording and use it across every agent in your workspace.

  • Sales and appointment calls

    Outbound qualification and reminder calls where a quick, natural reply keeps the caller on the line.

Other providers in this layer.

See all 12 text-to-speech providers
  • Custom TTS

    Text-to-speech

    Custom TTS lets your Bolna voice agent speak through your own OpenAI-compatible text-to-speech server, such as vLLM-Omni or an open-weight model, with your key.

  • Deepgram Aura

    Text-to-speech

    Deepgram Aura gives your Bolna voice agent fast, accurate English text-to-speech with the aura-2 model, and pairs with Deepgram transcription on one account.

  • ElevenLabs

    Text-to-speech

    ElevenLabs is Bolna's default voice. It gives your agent natural speech, Indian and American voices, voice cloning and models covering up to 74 languages.

Keep exploring the voice stack.

See where Cartesia fits in your workflow.

Walk through provider choice, stack trade-offs and the exact workflow you want Bolna to automate.

Ready for enterprise.

Secure, compliant and dependable, from the first pilot to full scale.

Contact Sales
  • 99.9% uptime

    Built for your busiest hour. Even if the model fails, the call doesn’t

  • Support and SLAs

    A dedicated team, with service levels shaped around your business

  • Security & compliance

    Data residency, encryption, access controls, full audit trails and more

  • Deploy in your cloud

    Run Bolna in your own cloud or on-premise, your data remains yours

Certified by

  • ISO 27001
  • DPDP compliant

Frequently Asked Questions

Use sonic-3.5. The Bolna docs recommend it for production agents. sonic-3 and Sonic 3.6 are also available, and all three work with every Cartesia voice in Bolna.