George Sterpu
George Sterpu
Hi @clarahohohoho `aus` stands for facial action units. We proposed to regress AUs from video representations jointly with the speech decoding task in order to overcome a learning issue of...
Hi @xjwla Are you reproducing the results from our ICMI'18 article on TCD-TIMIT? Yes, my training pipeline involves a multi-stage process where the same model is fine-tuned on gradually increasing...
Thanks a lot for the clarifications, @xjwla Hmm, I reckon that the audio sequence pre-processing could have a big impact on attention-based seq2seq models. The main difference between `logmel` and...
The example `run_audio.py` script is designed so that you can launch a full experiment under very similar conditions to what is described in the article, excepting the number of epochs...
Hi @xjwla Thanks for opening the issue. Could you please paste the error message here ? On my local copy of the TCD-TIMIT dataset I made a few corrections, most...
@clarahohohoho Yes, you would have to manually download any dataset you would like to use this project with. The script is only meant to serve as an example for writing...
@clarahohohoho Sorry, I am not aware of an alternative download page for TCD-TIMIT. I contacted the administrator of the webpage you mentioned. They are still working on finding a new...