Get started

Quickstart

Make your first speech synthesis request
1

Get your API key

  1. Sign up at platform.speechify.ai
  2. Go to API Keys
  3. Copy your default API key (or create a new one)

Set it as an environment variable so the SDKs pick it up automatically:

$export SPEECHIFY_API_KEY="your-api-key-here"

API keys are sensitive. Never expose them in client-side code or public repositories. See the Authentication guide for security best practices.

2

Install the SDK

$pip install speechify-api
Prefer raw HTTP? No install needed. Use the cURL tab in the examples below.
3

Generate speech

Send text to POST /v1/audio/speech. These examples are generated from our Fern SDKs and the API spec, so they switch languages and stay in sync with the live endpoint:

POST
/v1/audio/speech
1curl -X POST https://api.speechify.ai/v1/audio/speech \
2     -H "Authorization: Bearer <token>" \
3     -H "Content-Type: application/json" \
4     -d '{
5  "input": "Hello! This is the Speechify text-to-speech API.",
6  "voice_id": "geffen_32",
7  "audio_format": "mp3",
8  "model": "simba-3.2"
9}'
Try it

A successful call returns the audio payload:

Response
1{
2  "audio_data": "example",
3  "audio_format": "wav",
4  "billable_characters_count": 10,
5  "speech_marks": {
6    "chunks": [
7      {}
8    ],
9    "end": 1,
10    "end_time": 1,
11    "start": 1,
12    "start_time": 1,
13    "type": "example",
14    "value": "example"
15  }
16}
The Python and TypeScript SDKs return decoded audio bytes. The raw HTTP response base64-encodes the audio in the audio_data field, so decode it before saving.
4

Save and play

Assign the call above to response, then write the audio to output.mp3:

1with open("output.mp3", "wb") as f:
2    f.write(response.audio_data)

Then play it from the terminal:

$afplay output.mp3

Choose a voice

List the built-in voices to find one that fits, then pass its id as the voice_id:

GET
/v1/voices
1curl -G https://api.speechify.ai/v1/voices \
2     -H "Authorization: Bearer <token>" \
3     -d locale=en \
4     -d model=simba-3.2
Try it

Popular built-in voices: george, henry, carly, sabrina. You can also clone a voice from a short audio sample, with verified consent from the speaker.

For new integrations we recommend the streaming-native simba-3.2 model — set model: "simba-3.2" and pass one of its curated voices (beatrice_32, dominic_32, edmund_32, geffen_32, harper_32, hugh_32, imogen_32, wyatt_32) as the voice_id.

Add emotion

Use SSML to control how the voice sounds. Pass it as the input parameter and the API detects it automatically:

1<speak>
2  <speechify:style emotion="cheerful">
3    Great news! Your order has been shipped!
4  </speechify:style>
5</speak>
SSML also controls pitch, rate, pauses, and emphasis. See SSML and Emotion Control for the full reference.

Next steps

Stream audio

Process up to 20,000 characters with real-time audio streaming

Clone a voice

Create a custom voice from a short sample plus the speaker’s verified consent

Control speech

Use SSML for pitch, rate, pauses, and emphasis

API Reference

Full endpoint documentation with request/response schemas