AssemblyAI-TTS-1

AssemblyAI-TTS-1 is AssemblyAI's first text-to-speech model, built for phone voice agents: it streams audio in about 90 ms, offers 4,000+ voices and instant voice cloning, and costs $15 per million characters with unlimited concurrency and no annual contract.

Why it matters

For businesses and developers putting AI agents on their phone lines, natural-sounding, fast speech is what keeps callers from hanging up, and AssemblyAI is pricing it at about half what it says teams pay today, without concurrency tiers.

Highlights

• Generally available as of October 27, 2026, with audio streaming in about 90 ms in production; AssemblyAI reports a 47 ms median time to first byte in its own head-to-head test.
• More than 4,000 voices, plus instant voice cloning from an uploaded recording once you confirm you have the speaker's permission.
• Accents for English, Spanish, French, German, Italian and Portuguese, with more languages promised in the coming weeks.
• AssemblyAI compared it with Cartesia, Inworld and ElevenLabs models on four English test sets and published its methodology.
• Not yet available at launch: fine-tuned Pro voice cloning, custom pronunciation dictionaries and emotion tags.

Taken from the project's own README, docs or launch page. Claims about speed or accuracy are the maker's unless stated otherwise.

Worth knowing: The speed, quality and price comparisons are AssemblyAI's own. Voice cloning requires the speaker's permission.

Quick facts

Links

• Official site: https://www.assemblyai.com/blog/text-to-speech-api
• Docs: https://www.assemblyai.com/docs

Related ranking

See how established tools compare in Best Voice & Text-to-Speech. New releases aren't scored until we review them against the rubric.

Checked 2026-10-29 from the model's own pages. Prices and star counts change, so confirm on the official site. Listing here is not a ranking or an endorsement.