Voice API
    • Welcome to Async Voice API
    • Getting Started with the Async Voice API
    • Models
    • Billing & Pricing
    • API Reference
      • API Status
        • API Status Check
      • Text-to-Speech
        • Text to Speech (WebSocket)
        • Text to Speech
        • Text to Speech with Word Timestamps
        • Text to Speech (Stream)
      • Speech-to-Text
        • Speech to Text (WebSocket)
        • Speech to Text
      • Voice Management
        • Clone Voice
        • List Voices
        • Get Voice
        • Get Voices (Batch)
        • Update Voice
        • Delete Voice
        • Get Voice Preview
    • Advanced Guides
      • Embed Player
      • Custom Pronunciations
        • Embedding Custom Phonemes in Async Voice API
        • Pronouncing digits one‑by‑one
        • Insert Silent Pauses with <break>
    • Integrations
      • Integrate with Twilio
      • Pipecat Integration
      • Livekit Integration

    Welcome to Async Voice API

    Build natural-sounding voice applications with real-time performance.
    Async Voice API enables real‑time, low‑latency text‑to‑speech and speech-to-text for building voice assistants, chatbots, phone agents, and other audio experiences. The TTS API streams ultra‑realistic speech with about 300 ms latency, making it suitable for applications where responsiveness matters. It also supports voice cloning from just a 3-second sample and offers multi-language support.

    Text-to-Speech#

    Convert text into natural-sounding speech. Four delivery methods:
    MethodEndpointUse Case
    HTTPPOST /text_to_speechFull transcript → single audio file
    StreamingPOST /text_to_speech/streamingStart playback before generation completes
    TimestampsPOST /text_to_speech/with_timestampsAudio + per-word timing for captions
    WebSocketWSS /text_to_speech/websocket/wsIncremental text input (live LLM output, voice agents)

    Speech-to-Text#

    Transcribe audio into text. Two methods:
    MethodEndpointUse Case
    File UploadPOST /speech_to_text/Transcribe recorded audio files
    WebSocketWSS /speech_to_text/streamReal-time live transcription

    Models#

    Text-to-Speech#

    ModelTierDescription
    async_pro_v1.0ProHighest quality, English
    async_flash_v1.5StandardFast streaming, 6 languages
    async_flash_v1.0StandardBroad language support, 15 languages

    Speech-to-Text#

    ModelDescription
    async_asr_v1.0Multilingual speech recognition
    See the Models page for a detailed comparison.

    Key Features#

    Under 300ms latency — among the fastest TTS APIs available
    Voice cloning — create a custom voice from a 3-second audio sample
    15+ languages — with native pronunciation and text normalization
    Real-time streaming — WebSocket support for both TTS and STT
    Wallet billing — prepaid, pay-per-use, no subscriptions required

    Integrations#

    Async works with popular voice AI frameworks:
    Pipecat — open-source voice agent framework
    LiveKit — real-time audio/video infrastructure
    Twilio — calls, IVR, and contact centers
    n8n — workflow automation
    Picsart Flow — no-code AI workflows

    What's Coming Soon#

    Speech Enhancement (denoising, dereverberation)
    Speech-to-Speech (voice conversion)
    Modified at 2026-07-06 10:47:46
    Next
    Getting Started with the Async Voice API
    Built with