This is the official implementation of our paper [ENHANCING PRE-TRAINED SPEECH EMBEDDINGS FOR SPEECH RECOGNITION IN NOISY CONDITIONS"]. (Mohamed Nabih Ali, Daniele Falavigna, Alessio Brutti). The paper is under review in Computer Speech & Language Journal. The full code will be available as soon as the paper is accepted.
You can have a look at our repo. (https://github.com/mnabihali/Joint-training-embeddings) as a refrence.
