DOCSEARCHbyAlgolia
Submit Docs

AI TTS Microservice

website·Indexed 31 days ago

https://aitts.theproductivepixel.com/docs

A multi-provider text-to-speech platform for generating, streaming, and managing audio via REST API or MCP, providing unified access to services such as Google Cloud TTS.

Tokens
744
Snippets
4
Records
5
Agent score
0%

What's inside AI TTS Microservice

  1. Generate deep-links to specific voices in the Voice Gallery

    The /voices gallery supports URL filtering via query parameters to pin specific voices. The provider parameter is mandatory for deep-links. The voice parameter accepts three formats (resolved in priority order): 1. Full public voice_id (provider:lang-Family-Name), 2. Legacy canonical id (provider:voiceName), 3. Plain name (e.g., 'Kore'). The plain-name format is the most stable and is returned by the GET /api/v1/voices/{voice_id}/sample-url endpoint and the get_voice_sample_url MCP tool.

    Query Parameters:

    • provider: Provider slug
    • q: Fuzzy search text
    • tier: all, premium, or ultra
    • lang: BCP-47 language (e.g., en-US)
    • family: Model family slug (e.g., Wavenet, Gemini)
    • model: Sub-model selector (primarily for Gemini)
    • voice: The voice identifier
    # Plain-name (Preferred)
/voices?provider=google&voice=Kore&family=Gemini&lang=en-US&model=gemini-2.5-flash-tts

# Full public id
/voices?provider=google&voice=google:en-US-Wavenet-A

# Legacy canonical
/voices?provider=polly&voice=polly:en-US-Neural-Joanna

  2. Integrate AI TTS via Model Context Protocol (MCP)

    The platform provides MCP integration with two access methods: remote HTTP POST via /api/v1/mcp (57 tools) or a REST-proxied stdio implementation via npx @theproductivepixel/aittsm (56 tools). Available tools cover voice search, async and streaming generation, job management, shares, access codes, QR codes, library/storage management, and cost estimation. Authentication for MCP is handled via the AITTSM_API_KEY environment variable.
    export AITTSM_API_KEY=tts_YOUR_KEY
npx @theproductivepixel/aittsm

  3. Compare TTS provider capabilities and limits

    The service supports multiple providers with varying constraints:

    • Google Cloud TTS (Premium): Supports Chirp-HD, Chirp3-HD, Neural2, Studio, Wavenet, News, Polyglot. Supports SSML and bitrate config. Limit: 500,000 bytes.
    • Google Ultra (Gemini TTS): Supports 2.5 Flash, 2.5 Pro, 2.5 Flash Lite Preview, 3.1 Flash. Supports prompts and multi-speaker. Limit: 4,000 bytes text + 4,000 bytes prompt.
    • Amazon Polly (Premium/Ultra): Supports Generative, Neural, Long-Form, Standard. Supports SSML. Limit: 100,000 bytes.
    • Kokoro (Premium): Supports speed control and bitrate config. Limit: 5,000 bytes.
    • Azure AI Speech: Supports 11 public voice families. Async long-form input supports up to 500,000 UTF-8 bytes; short/stream input supports up to 4,000 bytes.

  4. Authenticate requests to AI TTS Microservice

    API requests require a Bearer token. API keys are created in the dashboard and start with the prefix tts_.
    Authorization: Bearer tts_YOUR_KEY

  5. Retrieve a shared voice collection via Voice Share Readback

    Use the public GET /api/v1/voice-shares/{code} endpoint to retrieve voice references associated with an 8-character base62 short code. This endpoint does not require authentication.
    // Response format
{
  "success": true,
  "data": {
    "code": "abc123xy",
    "voice_refs": [
      { "id": "google:en-US-Wavenet-A", "language": "en-US" }
    ],
    "voices": [],
    "created_at": "...",
    "expires_at": "..."
  }
}