0

I am trying to implement speech to text on my .net core 3.1 web app without using any cloud services like Azure, AWS, or Google.

I have found options like deepspeech, kaldi,pocketpheonix. It looks like they can be implemented on a .net core app but being a beginner in .net core and backend development I don't know how to use them and their documentation is not very beginner-friendly.

So if you can provide a brief overview of how to implement these apis on .net core or link to some articles, that would be very helpful.

Abid
  • 5
  • 1
  • `I am trying to implement speech to text on my .net core 3.1 web app` Would you like to recognize speech using browser user's device microphone on JavaScript client? – Fei Han Dec 30 '20 at 04:21
  • @FeiHan I have tried webspeech api but it's only supported by chrome. Do you know about any other solutions that is supported by more browsers? – Abid Dec 30 '20 at 14:36
  • Take a look at https://github.com/mozilla/DeepSpeech/tree/v0.9.3/native_client/dotnet and https://deepspeech.readthedocs.io/en/latest/DotNet-Examples.html#full-source-code. – Nishan Dec 30 '20 at 15:57
  • @Abid webspeech API in Chrome uses Google cloud services for that. – Olaf Dec 31 '20 at 09:18

1 Answers1

0

Just my personal view, some Kaldi or PocketSphinx experts might have a different view.

Kaldi + PocketSphinx

You will need to compile all sources yourself on Windows and Kaldi is a headache to build on Linux, so I guess it is not much better on Windows. That said, they use less resources and are usually quicker.

DeepSpeech

Just download the current English acoustic model (pbmm) and language model (scorer) and use the .NET example code. Should work without compilation.

Olaf
  • 158
  • 7