Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Creation of ASR api #1

Merged
merged 84 commits into from
Apr 24, 2020
Merged

Creation of ASR api #1

merged 84 commits into from
Apr 24, 2020

Conversation

ar13pit
Copy link
Contributor

@ar13pit ar13pit commented Mar 7, 2020

@ar13pit
Copy link
Contributor Author

ar13pit commented Mar 9, 2020

A basic framework for ASR API is ready. But somehow it seems to be slower than the script test_live.py.

@ar13pit
Copy link
Contributor Author

ar13pit commented Apr 7, 2020

@reinzor @LoyVanBeek can you guys look into the multiprocessing stuff I have in this PR? The design isn't very good. Although I have tested it thoroughly, I have observed latency issues. I adapted this from one of the test scripts I wrote testtest_live.py and there I do not have any object oriented design, so haven't faced any latency issues either.

src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
src/python/yapykaldi/asr.py Outdated Show resolved Hide resolved
.gitignore Show resolved Hide resolved
test/test_asr.py Outdated Show resolved Hide resolved
Logging functions take care of status printing
asr.stop is already in the interrupt handler, streamer.stop is being called by asr.stop
It is doing nothing here
@ar13pit
Copy link
Contributor Author

ar13pit commented Apr 22, 2020

The following things are currently broken:

  1. Using an audio file streamer for ASR. (Works for python2 if the futurize lines are removed from test_asr.py)
  2. Writing audio file upon completion of recognition in python3

@ar13pit
Copy link
Contributor Author

ar13pit commented Apr 23, 2020

@LoyVanBeek A new issue spotted:

When using the WaveFileSource streamer, the stream misses out on probably the last chunk. You can test this out with the audio file at data/lsen1.wav relative to the repo root.

@ar13pit ar13pit requested a review from LoyVanBeek April 23, 2020 22:31
@ar13pit
Copy link
Contributor Author

ar13pit commented Apr 23, 2020

@LoyVanBeek This is ready now. Have a look before I merge this.

@ar13pit ar13pit merged commit 519e76e into master Apr 24, 2020
@ar13pit ar13pit deleted the feature/asr_api branch April 24, 2020 12:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Add api for continuous speech recognition Create yapykaldi API
3 participants