Skip to content

qianhwan/speech-command-classifier

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

speech-command-classifier

Build docker

docker built -t speech-command .
docker run -it -p 8888:8888 \
    -v /home/qianhui/Dataset/speech_commands_v0.01/:/data
    -v $PWD:/home/workspace
    speech-command /bin/bash 

Preprocess dataset

python bin/preprocess.py \
    --data-path /data/ \
    --output-path metadata/

This will create three metadata csv files (for train, validation and test respectively) inside OUT-DIR, each csv file has two columns: audio file path and label.

Train

Use the following command to train model, make sure you run preprocess.py first.

python bin/train.py \
    --train-metadata metadata/metadata_train.csv \
    --validation-metadata metadata/metadata_val.csv \
    --config configs/config.yaml \
    --output outputs/

Use --resume to continue training from a previously saved checkpoint.

Monitor training and validation loss

mlflow ui

Predict

python bin/predict.py \
    --audio /data/cat/0819edb0_nohash_0.wav \
    --model models/best-accuracy-229896.pth \
    --config configs/config.yaml

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors