HiFi++: a Unified Framework for Bandwidth Extension and Speech Enhancement
This work addresses audio quality improvement for speech processing applications, presenting an incremental advancement by adapting existing vocoder technology to new tasks.
The paper tackled bandwidth extension and speech enhancement by proposing HiFi++, a unified framework based on generative adversarial networks, achieving performance comparable to or better than state-of-the-art methods with significantly reduced computational resources.
Generative adversarial networks have recently demonstrated outstanding performance in neural vocoding outperforming best autoregressive and flow-based models. In this paper, we show that this success can be extended to other tasks of conditional audio generation. In particular, building upon HiFi vocoders, we propose a novel HiFi++ general framework for bandwidth extension and speech enhancement. We show that with the improved generator architecture, HiFi++ performs better or comparably with the state-of-the-art in these tasks while spending significantly less computational resources. The effectiveness of our approach is validated through a series of extensive experiments.