Voice and music · by Microsoft
Azure Voices
Narrate long audiobooks in hundreds of voices across 100+ languages.
text → speech · speech → profile · text, document → speech
- Way in
- Official APISign up yourself, api key
- To start
- Free tierF0: 0.5M neural chars/month
- Runs
- In the cloudAlso through Azure
- Checked
- Nov 3, 2026Prices as of that day
What it can do 5
Synthesize speech with neural or HD voices (SSML)text → speech$15/1M chars Neural; $22/1M Neural HD
Train and host a professional custom neural voiceLimited Access approvalspeech → profiletraining $52/compute-h; hosting $4.03/h; synthesis $24/1M ($48 HD)
Create a personal voice from a short consented samplepre-approved use cases onlyspeech → profilecreation free; synthesis per 1M chars
Synthesize long audio (audiobooks) as a batch jobtext, document → speech$100/1M chars
Run speech-to-speech voice agents (Voice Live API: gpt-realtime-2.1, phi4-mm-realtime, azure-realtime)speech → speech, video, eventsper 1M tokens, tiered Pro/Standard/Lite/BYO by chosen LLM (see pricing)
Does the same job
Prices and limits as of Nov 3, 2026, from Microsoft’s own pages. They change; check before you rely on them.Not affiliated with Microsoft. Not advice.Something wrong?