ModelsSep 2, 2026 · 01:516 sources

Meta Launches First Real Time Audio Perception Model Muse Voice Transcribe for $3 per 1,000 Minutes

From huggingnews.com

Image: huggingnews.com
AI;DR

Meta Superintelligence Labs released Muse Voice Transcribe, a multimodal LLM that combines streaming automatic speech recognition and speaker diarization for $3 per 1,000 audio minutes. The model processes audio in 80ms chunks, supports more than 70 languages, and can manage hour long sessions with over 20 speakers. It currently powers voice input in Muse Code and dictation in the Meta desktop app.

Read the sourceSign in to suggest

Sources

+ 3 more
  1. SUPPORT@aiatmeta“Audio is processed in 80ms chunks (12.5 Hz), one token each”x.com
  2. SUPPORT@finkd“trained across 70+ languages (with 25 validated at launch)”x.com
  3. SUPPORT@aiatmeta“achieves the pareto frontier on speed-accuracy trade-off measured by time to final transcription”x.com
#meta#llm#inference#benchmarkMarkdown
Content history (1)
  • 2026-10-04 · Summary · vi · Wording fix
    Meta Superintelligence Labs vừa phát hành Muse Voice Transcribe, một LLM đa phương thức k…
    Meta Superintelligence Labs vừa phát hành Muse Voice Transcribe, một LLM đa phương thức k…

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today