The Text-to-Speech (TTS) API converts plain text into synthesized speech. Fast mode accepts the complete text in one HTTPS POST and streams audio in the same response, so your application can start playback while receiving it.
Use the TTS API for interactive voice response systems, notifications, accessibility experiences, and short-form narration.
Key features
- Converts up to 4,000 characters of plain text per request
- Streams audio in the response, so playback can start before synthesis finishes
- Offers ten voices
- Controls speaking speed and volume
- Returns raw 44.1 kHz mono PCM audio
How it works
- Get AI Services credentials.
- Send the text in the
inputfield toPOST /aiservices/tts/synthesize. - Optionally set
voice,speed,volume, andresponse_formatin the same JSON object. - Read the response body as raw PCM audio until the stream ends.
For request fields, examples, and error handling, see Fast mode.
Language and voices
The API supports US English (en-US) input. There is no language field, and a voice's accent doesn't add support for other input languages.
Set voice to one of these ten voice IDs. The default is jenna.
| Voice | Voice ID | Accent | Character |
|---|---|---|---|
| Alex | alex | American English | Charming and friendly |
| Ariel | ariel | American English | Professional and clear |
| Ava | ava | American English | Calm and empathetic |
| Brooke | brooke | American English | Calm and professional |
| Hazel | hazel | American English | Enthusiastic and warm |
| Jenna | jenna | American English | Clear and smooth |
| Mark | mark | American English | Professional and empathetic |
| Reid | reid | American English | Calm and crisp |
| Scott | scott | American English | Friendly and deep |
| Tony | tony | British English | Charming and eloquent |
Voice IDs are case-insensitive. An unsupported voice returns 400 UNSUPPORTED_VOICE; the service never substitutes another voice.
Audio format and limits
| Setting | Value |
|---|---|
| Response format | pcm_44100 |
| Encoding | Signed 16-bit little-endian PCM |
| Sample rate | 44,100 Hz |
| Channels | Mono |
| Container | None; no WAV or RIFF header |
| Maximum input length | 4,000 Unicode code points after JSON decoding |
| Maximum request body | 32 KiB (32,768 bytes) |
The API accepts plain text only. It doesn't support SSML, streaming text input, batch synthesis, model selection, or other output formats.
Resources
AI Services are part of the Zoom Build platform. Use Zoom.ai to discover, purchase, build with, and manage AI Services products. You can also create projects, get credentials, start a free trial, purchase a plan, and monitor usage.