This repository contains the implementation of (MQGAN) for audio synthesis. The project is structured to facilitate the entire workflow from data preparation to model deployment.

Abstract: In this work, we propose CleanMel, a single-channel Mel-spectrogram denoising and dereverberation network for improving both speech quality and automatic speech recognition (ASR) performance ...

Abstract: Mel power spectrogram has been extensively used as audio pre-processing for both feature extraction and transformation. Between many, one of the most used libraries is Librosa. In this paper ...

Diffusion Speech is a diffusion-based text-to-speech model. Our speech synthesis pipeline is quite simple. We use a diffusion transformer model (DiT) to predict the duration of each phoneme. Then we ...