Meta releases real-time audio model for AI assistants
meta releases muse voice transcribe, a real-time audio model that transcribes speech for ai assistants, at $0.18 per hour, supporting 70+ languages.
meta has released muse voice transcribe, a real-time audio model that transcribes speech, detects sentence boundaries, and identifies up to 20 speakers without separate systems. the model breaks audio into 80-millisecond chunks and adjusts delay for each word based on difficulty, balancing speed and accuracy. it is priced at $0.18 per hour, undercutting competitors like openai and elevenlabs, and supports over 70 languages. muse voice transcribe is available in meta ai and through the meta model api, and is seen as a foundation for ai assistants that can listen in on real conversations.