Microsoft AI releases MAI-Transcribe-2-Streaming, ranked No. 1 for accuracy on AA-WER Streaming

ArtificialAnlys says Microsoft AI’s new speech-to-text model ranks first for both final- and first-partial-transcript accuracy on AA-WER Streaming. It reports a 2.5% word error rate for final transcripts delivered 0.13 seconds after speech ends, compared with 2.7% and 0.49 seconds for Grok Voice Transcribe 2.0. Streaming transcription costs $0.54 per hour of audio, it says.