https://github.com/Arcadia2142/voice-recognize

https://github.com/Arcadia2142/voice-recognize

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/Arcadia2142/voice-recognize
• Owner: Arcadia2142
• Created: 2022-01-29T16:19:08.000Z (almost 3 years ago)
• Default Branch: master
• Last Pushed: 2022-01-31T06:34:07.000Z (almost 3 years ago)
• Last Synced: 2024-06-27T13:30:50.298Z (5 months ago)
• Language: Python
• Size: 40 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README

Awesome Lists containing this project

README

        
Now this is just dev project.

For install type: make

Project is based on DeepSpeech:

https://deepspeech.readthedocs.io/en/r0.9/

Voice recognize run by(after install): voice_recognize.sh

For acceleration on GPU you need linux with nvidia driver 418+ and docker support..:

https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html

For train from scratch (GPU-only) download train data from https://commonvoice.mozilla.org/en/datasets

to ./train/ add your alphabet.txt use ./train/prepare_lang.sh  for prepare data

and train model by full_learn.sh 

Czech DeepSpeach model is from repository:

https://github.com/comodoro/deepspeech-cs