PRODUCTION SCENARIO
A support line needs the caller's spoken request turned into text inside the app as the caller speaks, with partial text appearing before the caller finishes the sentence.
Which Azure Speech capability should the team use?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Real-time transcription
Real-time transcription gives 'instant transcription with intermediate results for live audio inputs', which is what shows partial text while the caller is still speaking. Fast transcription is synchronous and faster than real time, but it works on stored audio files and returns only the finished result. Batch transcription processes files asynchronously. Speech translation transcribes and translates, and no second language is wanted.