Text to speech
MAI-Voice-2.1-Flash
MAI-Voice-2.1-Flash is a text-to-speech model built for fast, low-latency generation. It produces high- fidelity, natural, and expressive speech across 23 languages and supports gated instant voice cloning, all while being optimized for real-time responsiveness. Its human-like intonation, rhythm, and emotional nuance make it ideal for voice agents, assistants, and other interactive scenarios where latency and cost are critical.
MP3, WAV, OPUS, PCM, AAC, FLAC · 4 voices · Up to 4,096 characters
$15.75Per 1M characters
Price
- Per 1M characters
- $15.75
Includes the MiniRouter fee. Tokens are not charged.
Estimate your cost
Priced per character of input text.
44 / 4,096 characters
Estimated cost
Includes the MiniRouter fee.
Estimate only; nothing is sent. Your charge settles to the provider’s reported characters plus the MiniRouter fee.
Use the Speech API
Set MINIROUTER_API_KEY to your inference key and copy the example.
curl --fail-with-body --retry 0 --max-time 120 \
https://api.minirouter.sh/v1/audio/speech \
-H "Authorization: Bearer $MINIROUTER_API_KEY" \
-H 'Content-Type: application/json' \
-D response.headers --output speech.mp3 \
--data '{
"model": "microsoft/mai-voice-2.1-flash",
"input": "The quick brown fox jumps over the lazy dog.",
"voice": "en-US-Harper",
"response_format": "mp3"
}'Keep the request ID and check Request logs for the settled cost. A failed or disconnected request may still incur a cost; these examples disable automatic retries.
How it works
- Endpoint
- POST /v1/audio/speech
- OpenAI speech body: model, input, voice, response_format, speed, instructions. Raw audio bytes come back.
- Limits
- 4,096 characters
- Speed 0.25–4.0. No streaming.
- Formats
- mp3, wav, opus, pcm, aac, flac
- mp3 by default.
- Voices
- en-US-Harper, en-US-Nova, en-US-Davis, en-US-Jenny
- Pass one as voice.