Automatic Speech Recognition for Gujarati Language
Code to Train a Model for gujarati text to speech using the Microsoft Research Open Data
https://msropendata.com/datasets/7230b4b1-912d-400e-be58-f84e0512985e
This Model can be used to recognize any language as long the necessary changes are made to Corpora.py file.
MOST OF THE DIRECTORY LINKS ARE ABSOLUTE TO MY COMPUTER AND NEED TO BE CHANGED
Setup
Ensure that the Following File Structure is present
-Root
---gu-in-Test
-------/transcription.csv
-------/Audios
---gu-in-Train
-------/Audios
-------/transcription.csv
Initialize
python read_dataset.py
Training
python model_train.py
Testing
python test_model.py