Errors with using transcribe.py

DeepSpeech
#1

Can someone please tell me how to use Deepspeech’s transcribe.py?

I created a venv with python3.6 and cloned Deepspeech (v0.6.0) into my folder. I’m trying to run transcribe.py but no matter what commands I use or how many additional libraries/packages I install, I’m still not able to run it. I’m sorry if this is vague but could anyone point me to a resource that can show me how this will work? The errors I’m receiving are similar to this: Error while trying to transcrible wav file with transcribe.py

Please kindly help. I’ve tried my best to read the docs as much as possible, but I just can’t seem to find any resource that can get this file to work.

#2

This project is not what it used to be, head over to coqui.ai instead. You’ll find help there.

#3

I see. Thank you for that, but is transcribe.py still usable? I have been working on a project on this for so long and i would really appreciate if there are resources for this too please

#4

To be more specific, this is my issue now

This is the command i’m using:

python3 transcribe.py --src sample_audio_file.wav --scorer deepspeech-0.9.3-models.scorer

output:
I Could not find best validating checkpoint.
I Could not find most recent checkpoint.
E All initialization methods failed (['best', 'last']).

Please help :frowning:

#5

Have you asked in their gitter yet?

1 Like