question

Hi-6046 avatar image
1 Vote"
Hi-6046 asked ramr-msft answered

Speech recognition text format

I have started to use Azure Cognitive services for Speech to Text API.

The quality of results are really good but the output is a monolithic bloc of text without any paragraph or newline.

I don’t see a way in the doc to get a well formatted output. Is that not possible ?

Thanks Mike

azure-cognitive-services
· 3
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

@Hi-6046 Thanks for the question. Can you please add more details about the version of STT API and output that you are getting. Also Please confirm are you using the STT API Batch Transcription API or Speech-to-text REST API for short audio.


0 Votes 0 ·

Hi,

Thanks a lot for looking at my question.
I am using STT API 3.0 ( endpoint : https://southcentralus.api.cognitive.microsoft.com/speechtotext/v3.0/transcriptions)

I am using the API Batch Transcription API since I am working with audio files.
I am then retrieving the JSON results and more specifically the property "display" from "combinedRecognizedPhrases".

The results are great but the format is not present. I just get a monolithic bloc of text.

Is there a way to get a formatted output with paragraphs, new line ?

Thanks a lot!
Mike

0 Votes 0 ·

@Hi-6046 Thanks for the details. We are checking with the product team and will update on the same.

0 Votes 0 ·

1 Answer

ramr-msft avatar image
0 Votes"
ramr-msft answered

@Hi-6046 Thanks for the details. There is currently no formatting supported for paragraphs. You could use the single phrase results, It is not exactly a paragraph but might be better readable.

5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.