AssemblyAI launches first TTS model on LiveKit: phone-call-trained, 90ms first audio, $15/1M chars
AssemblyAI · x · 2026-11-04
AssemblyAI's first TTS model, built for voice agents and trained on real phone conversations, is now available on LiveKit Inference. It delivers first audio in 90ms, offers 4,000+ voices across 6 languages, includes word timestamps, and costs $15 per 1M characters.
Related event: AssemblyAI launches first TTS model with 90ms first-audio latency(2 posts)→
More from Multimodal
- Suno song mourns AI-automated economists in Russian alt-rock — shockingly good — aniketapanjwani · 2026-11-04
- Fully Local AI Short Film Rendered on a Mac, Workflow Open-Sourced as Phosphene — Opening-Ad5541 · 2026-11-04
- Excel Shonen Anime Returns with Episode 2: 'God of Death' — superfatbeagle · 2026-11-04
- AI-generated cat 'uses up 9 lives' on a 15-second vacation in Dreamina video — umesh_ai · 2026-11-04
- YC-backed Vyra launches conversational AI agent that edits raw footage into short videos — ycombinator · 2026-11-04
- Hala Recreates Haraj's Ad Video with MCP + Claude, Generating Motion Graphics with Voiceover — aziz4ai · 2026-11-04