Meta's new real-time audio model is the foundation for

What happened

Meta's Superintelligence Labs have released their first real-time audio perception model. It transcribes speech, tells speakers apart, and detects sentence boundaries during live conversation. The Spark-family model breaks incoming audio into 80-millisecond chunks. After each one, it decides whether to keep listening or output the next word as text, controlling how much context it gathers before committing to a transcription. More waiting means better accuracy but longer delays.

0 percent, depending on whether the model detects utterance endings itself or relies on an external system. Competition is fierce. OpenAI released GPT-Realtime-Whisper for the same purpose in May and cut transcription model prices in July. " In a staged demo, Meta employees argue that reliable speech recognition is the foundation for personal AI agents that listen in on real conversations through AI glasses.

Why it matters

Muse Voice Transcribe adjusts the wait time dynamically for each word based on difficulty, according to Meta. Easy words come out faster, harder ones get more listening time. Meta trained this behavior with reinforcement learning, rewarding the model for low error rates and short delays at the same time. Meta builds the remaining features into the same model, without separate systems.

For speaker attribution, the model marks speaker changes in the running text and tags each passage with an identifier from A to Z. For sentence boundaries, it marks where each utterance begins and ends. Both tasks are trained jointly with the speech recognition. The model can tell more than 20 speakers apart at once and handle recordings over an hour long without post-processing.

In a demo with eight people in a room, the system assigns words to individual speakers in real time, according to Meta. Meta says Muse Voice Transcribe was trained on more than 70 languages, with 25 tested in depth. The model also handles code-switching, where speakers jump between two languages mid-sentence. " Artificial Analysis confirms Meta's claims in an independent evaluation. 16 seconds after a speaker finishes talking. 0 percent.

What to watch

In Germany, a ban on Meta's camera glasses was recently discussed, though Germany's Federal Network Agency decided not to pursue it. Muse Voice Transcribe now powers voice dictation in Meta AI and Muse Code and is available through the Meta Model API. Users can try it by holding the "Fn" key in any app. Meta undercuts the field on pricing. 50 each.

2, where Meta competes on price rather than peak performance. The company hasn't disclosed the model's parameter count, training data volume, or audio data sources, and it isn't releasing the weights. Meta reorganized its AI division in the summer of 2025 under the Superintelligence Labs umbrella, recruiting top researchers from OpenAI, Google DeepMind, and Apple with pay packages up to $300 million over four years. Not all of them stayed. Some went back to OpenAI after just a few weeks.