AssemblyAI launches STT-3.7 and its first TTS model for real-time voice agents
AssemblyAI · x · 2026-11-20
AssemblyAI has launched AssemblyAI-STT-3.7 (real-time speech-to-text) and AssemblyAI-TTS-1, the company's first text-to-speech model. TTS-1 is built for voice agents on live calls, emphasizing natural sound and fast response to keep callers engaged. In an accompanying huddle, real-time models engineer Ilya Sklyar discusses the biggest change in STT-3.7: how a voice agent can ignore background noise like a running TV during a call. The platform covers 99 languages, plus pre-recorded/realtime/sync transcription, speech understanding, guardrails, an LLM gateway, and self-hosting.
Related event: AssemblyAI Launches STT-3.7: Three Specialized Speech-to-Text Models(6 posts)→
More from Multimodal
- SenseNova U1 Pro debuts at No.6 on DesignArena image leaderboards, behind only OpenAI — alifcoder · 2026-11-20
- Turning Sonnet 3's glossolalia into hyperpop songs with Suno — repligate · 2026-11-20
- Dev builds pixel arcade shooter with Astra 6.2 and Scenario in a few rounds — AIandDesign · 2026-11-20
- AI cover of 'Kiss from a Rose' impresses past the first seconds of jank — rickasaurus · 2026-11-20
- 9 MOSS-Image Turbo LoRAs vs 50-step baseline: classical painting texture showdown — darkside1977 · 2026-11-20
- Pure-vision retrieval tops ViDoRe V3 leaderboard: GLM agent + EVIE hits 70.01 nDCG@10 — CShorten30 · 2026-11-20