Working online speech recognition based on RNN Transducer. ( Trained model release available in release )
After unzipping the files, move all folders to the root path of this project
Your current project should look like this:
logs/ <--- log file
rnnt-
BPE-{token size} <-- your unzip file
models/
models.py
If you are trying out the English speech recognition, you can try to execute this command
python stream.py --flagfile ./flagfiles/E6D2_LARGE_Batch.txt
--name rnnt-m-bpe
--model_name english_43_medium.pt
--path 3729-6852-0035.flac