Repository for experimenting with different CTC based model designs for ASR. Supports live recording and testing of speech and quickly creates customised datasets using own-voice dataset creation scripts!
- Recommended > use virtualenv installed with python2.7 (3.x untested and will not work with Core ML)
git clone https://github.com/mlrobsmt/KerasDeepSpeech
pip install -r requirements.txt
- Get the data using the import/download scripts in the folder, LibriSpeech is a good example.
- Download the language model (large file) run
./lm/get_lm.sh
- To Train, simply run
python run-train.py
In order to specify training/validation files usepython run-train.py --train_files <csvfile> --valid_files <csvfile>
(see run-train for complete arguments list) - To Test, run
python run-test.py --test_files <datacsvfile>
- Mozilla DeepSpeech
- Baidu DS1 & DS2 papers