ESPNet-Model-Inference

ESPnet Model Bulk Speech Transcribe - Python

Steps

1. Install ESPnet as instructed here.
2. From the espnet root directory, run cd egs/librispeech/asr1 && . ./path.sh
3. To run bulk inference: python bulk_inference.py wav_folder (Will dump JSONs in decode folder)
4. To get txt transcripts, set output folder in esp2txt.py and run python esp2txt.py

Reference

• ASR Demo