Speech Generation

Generate speech from text using a saved voice (voice_id) or a one-off reference audio clip (ref_audio).

Generate

Generate speech using a previously saved voice (voice_id) identifier.

import base64
from pathlib import Path
from mistralai.client import Mistral

client = Mistral(api_key="your-api-key")

response = client.audio.speech.complete(
    model="voxtral-mini-tts-2603",
    input="Hello! This is Voxtral, Mistral's text-to-speech model.",
    voice_id="your-voice-id",
    response_format="mp3",
)

Path("output.mp3").write_bytes(base64.b64decode(response.audio_data))
print("Saved to output.mp3")

import base64
from pathlib import Path
from mistralai.client import Mistral

client = Mistral(api_key="your-api-key")

response = client.audio.speech.complete(
    model="voxtral-mini-tts-2603",
    input="Hello! This is Voxtral, Mistral's text-to-speech model.",
    voice_id="your-voice-id",
    response_format="mp3",
)

Path("output.mp3").write_bytes(base64.b64decode(response.audio_data))
print("Saved to output.mp3")

Response Audio Formats

Format	Description
`mp3`	Compressed, suitable for most use cases
`wav`	Uncompressed PCM, highest quality
`pcm`	Raw float32 LE samples — recommended for streaming (lowest latency)
`flac`	Lossless compression
`opus`	Low bitrate, good for streaming