Microsoft launches MAI-Transcribe-2-Streaming
Microsoft’s new streaming transcription model is built to return usable partial text almost immediately, so voice agents can act before a speaker finishes.
MAI-Transcribe-2-Streaming supports 60 languages with continuous automatic language detection. Microsoft says it begins producing partial hypotheses just over 100 milliseconds after receiving audio, then revises and commits stable text as more context arrives. The company claims it ranks first on Artificial Analysis for final and partial transcript accuracy, and that its internal tests showed words appearing twice as fast as its closest competitor. The introductory price is $0.54 per audio hour through year-end. TestingCatalog's note
MAI-Transcribe-2-Streaming supports 60 languages with continuous automatic language detection. Microsoft says it begins producing partial hypotheses just over 100 milliseconds after receiving audio, then revises and commits stable text as more context arrives. The company claims it ranks first on Artificial Analysis for final and partial transcript accuracy, and that its internal tests showed words appearing twice as fast as its closest competitor. The introductory price is $0.54 per audio hour through year-end. TestingCatalog's note
score 5