Fish Audio releases the Live 3 voice model
The story so far
Fish Audio has released the full-duplex voice model Live 3, which can perform both listening and speaking and learn real-time dialogue timing, with the fastest response time for short responses (such as 'mm-hmm') being 186 milliseconds. Live 3 can be deployed on the Fish Agent platform, API, or on-premises enterprise, currently only supporting English, priced at 0.035 USD/minute (until December 31, 2026).
AI-written from the reports · Updated 3 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- MarkTechPostFish Audio releases Live 3, a full-duplex voice model that can both listen and speak
Fish Audio releases full-duplex speech model Live 3, capable of hearing and learning real-time dialogue timing simultaneously, with the fastest response time of 186 milliseconds; deployable on Fish Agent platform, API, or on-premises enterprise, English supported, priced at $0.035 per minute (until December 31, 2026).
Heat over time
- Now
- 3
- Peak
- 1011 Nov, 14:00
- 24h change
- -49%
Hover or tap the chart to see each hour; use the arrow keys on a keyboard.