Skip to content

slavachernyshev/whisper-diarization

 
 

Repository files navigation

Whisper Diarization

Арендовать север можно на immers-cloud. При этом стоит исопльзовать конфигурацию ubuntu 22.04 с CUDA 12.6. Такая конфигурация попросит 120 Гб диск.

Commands on local computed:

# Add envs
export CERT_PATH=~/Downloads/trans-205936-slavacher.pem
export GPU_IP=81.94.150.14

# Connecting
$ ssh -i ${CERT_PATH} -p 22 ubuntu@${GPU_IP}

# Copying file
$ scp -i ${CERT_PATH} -P 22 audio.wav ubuntu@${GPU_IP}:~

# Copying file back
$ scp -i ${CERT_PATH} -P 22 ubuntu@81.94.150.14:~/audio.txt .
$ scp -i ${CERT_PATH} -P 22 ubuntu@81.94.150.14:~/audio.csv .
$ scp -i ${CERT_PATH} -P 22 ubuntu@81.94.150.14:~/audio.srt .

Commands on remote host:

$ sudo chmod +x ./run.sh && ./run.sh

About

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 54.4%
  • Python 44.9%
  • Shell 0.7%