Microsoft provides Word with transcription functionality for English-language audio
The online version of Microsoft Word will have a transcription function that converts English-language audio into text. It is possible to upload and convert a previously recorded audio file, but speech can also be recorded directly from Word.
The transcription function in Word can recognize different speakers, according to Microsoft, so that the text conversion also directly distinguishes between them. Parts of the recording can be listened to via timestamps and it is also possible to adjust the transcription if the conversion did not go completely smoothly. The text is visible in Word on the right side of the main document. From there, the text is easy to insert into the open document. Word stores recorded audio in the wav file format on OneDrive, writes The Verge.
Uploaded audio files can only be converted if they are in MP3, WAV, M4A, or MP4 format. There is a limitation here: for previously recorded files that are uploaded in Word for conversion to text, there is a limit of five hours per month. And each upload cannot exceed 200MB. These restrictions do not apply to audio that is recorded directly into Word, ie no file is uploaded first.
The feature will be available in Word for the web starting Wednesday and is open to all Microsoft 365 subscribers. The feature is supported in Edge and Chrome browsers and is currently only usable with English-language speech, although more languages are under development. A Transcribe in Office feature will be available by the end of the year, according to Microsoft, so Android and iOS users will be able to use the feature over time.