> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://space.respeecher.com/docs/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://space.respeecher.com/docs/_mcp/server.

# Limits

The following limits apply to API usage.
[Contact us](https://www.respeecher.com/contact) if you need to extend them.

* Maximum of 5 concurrent text-to-speech generations across all endpoints for one account.
  (Maximum of 1 for accounts on free trial.)
* Maximum response size of 100 MB for the [Bytes](./api/tts/bytes) endpoint.
  Audio files will be silently truncated above this threshold.
  Use multiple calls to [Bytes](./api/tts/bytes), or use the streaming endpoints,
  [SSE](./api/tts/sse) and [WebSocket](./api/tts/web-socket),
  if you need longer audio files.
* Maximum transcript length of approximately **5,000 characters** for the [Bytes](./api/tts/bytes) endpoint.
  The endpoint operates under a 120-second HTTP request timeout, which limits how much audio can be generated in a single call.
  If your transcript exceeds 5,000 characters, use the [SSE](./api/tts/sse) or [WebSocket](./api/tts/web-socket) endpoints instead.