MAI-Transcribe-2-Streaming
ClosedAudioAudio
Microsoft's MAI-Transcribe-2-Streaming, a real-time speech-to-text model for streaming audio transcription.
—context
—per 1M tokens (in / out)
—parameters
8stories in our feed
Specs
hover a value to see its sourceContext window—
Max output—
Parameters—
Knowledge cutoff—
Licence—
Commercial useUnknown
Input—
Output—
Familymai-transcribe
API price
USD per 1M tokensNo public API price yet.
2026-10-04
2026-10-01
Microsoft发布MAI语音转录与生成模型TestingCatalog
微软发布三款语音模型Microsoft AI
微软发布MAI-Transcribe-2-Streaming实时转录模型Mustafa Suleyman