Skip to content

gotois/vosk

Repository files navigation

Vosk Speech Recognition

Installation

.env

VOSK_MODEL_PATH=/models/vosk-model-small-ru-0.22
bash install.sh
docker-compose up --build

Supported Audio

  • Format: WAV
  • Encoding: PCM-16 LE
  • Depth: 24-bit
  • Sample Rate: 16 kHz

Example

curl -X POST -F "audio=@example.wav" http://127.0.0.1:2700/recognize | jq

About

Russian speech recognize

Resources

License

Stars

1 star

Watchers

0 watching

Forks

Contributors