Meta's new real-time audio model is the foundation for AI assistants that never stop listening

Meta has released Muse Voice Transcribe, an audio model capable of streaming transcription with 80-millisecond latency. By accurately identifying speakers and sentence breaks in real time, the technology provides a technical foundation for voice-based AI assistants that maintain continuous, fluid conversations. Its performance in processing speed and accuracy aims to reduce the delays currently associated with speech-to-text interactions.