Great work!
#1
by jpgallegoar - opened
Hey moxeeeem, I appreciate this model, great job! Would it be possible to write some information on how this was trained? Especially, what dataset did you use, what training pipeline? I'm interested in finetuning it further for spanish correction. Thank you in advance.
Hi Juan Pablo! Thanks for your attention to the project.
You can find the code for training the model here: https://github.com/moxeeem/ASR-pronunciation-correction/blob/main/analysis/finetune_wav2vec2.ipynb
The repository also contains a detailed description of the dataset compilation.
At the moment, everything is, unfortunately, in Russian, but we plan to continue working on the project soon to supplement and correct shortcomings on github and huggingface