Meta Muse Voice Transcribe Leads Speech-to-Text Benchmark

The New Stack posted that Meta's Muse Voice Transcribe leads a streaming speech-to-text benchmark. The post states the model tracks more than 20 speakers and learns via reinforcement learning how long to listen before each word. Their linked article claims Meta beat OpenAI and Google at real-time transcription. All details come from the single tweet and source link supplied in the packet. No independent confirmation appears in the evidence.