Microsoft describes long-audio transcription with speaker diarization
An article published on March 2, 2026 describes a model that Microsoft says transcribes up to 60 minutes of audio in one pass and supports more than 50 languages.
Source: VibeVoice-ASR da Microsoft oferece transcrição de áudio longo (x.com). Text prepared with AI from this source.
What happened and what to do
In an article published on March 2, 2026, Microsoft says its speech-recognition model transcribes audio up to 60 minutes long in a single pass. The system includes speaker diarization, timestamps and hotword support, as well as more than 50 languages without requiring language configuration.
A company processing recordings could assess a pipeline to receive files, transcribe them and store text, timings and speaker labels for retrieval. An implementation could include failure monitoring, access controls and a dashboard to track processing volume and status. The right design depends on the case’s privacy, quality and operational requirements.
How the consultancy can help
Wendelmaques can assess a recording workflow and its privacy requirements, define a scoped implementation, and support operation of the transcription and monitoring pipeline.
Next step
Send a short description of your case to receive a scoped proposal for diagnosis and implementation.
Let's assess your project
I reply with scope and a proposal.