What was announced?
Microsoft introduced its first streaming transcription model. According to the company's announcement, the new models focus on faster transcription and multilingual voices.
What does streaming transcription mean?
Streaming transcription means producing text piece by piece while speech is still ongoing, instead of waiting for the whole audio to be recorded and processed. This approach aims to reduce latency in use cases such as live captions and real-time note taking.
Multilingual voices
The announcement is not limited to transcription; it also covers multilingual voices. Microsoft shared these models through its official source.
Details remain limited
Information shared so far is limited in terms of model names, technical specifications and access conditions. This story will be updated as more details emerge.
