We present a novel high-fidelity real-time neural vocoder called VocGAN. A recently
developed GAN-based vocoder, MelGAN, produces speech waveforms in real-time …

The wideband or super wideband speech is one of the most prominent features in real-time
communication services, with higher resolution spectrum. However, it requires higher …

In recent years, neural vocoders have surpassed classical speech generation approaches in
naturalness and perceptual quality of the synthesized speech. Computationally heavy …

Deep-learning based methods have shown their advantages in audio coding over traditional
ones but limited attention has been paid on real-time communications (RTC). This paper …

Low bitrate speech codecs have become an area of intense research. Traditional speech
codecs, which use signal processing methods to encode and decode speech, often suffer …

Recently, a conventional neural decoder for speech codec has shown promising
performance. However, it typically requires some prior knowledge of decoding such as bit …

End-to-end speech coding models achieve high coding gains by learning compact yet
expressive features and a powerful decoder in a single network. A challenging problem as …

Recently end-to-end neural audio/speech coding has shown its great potential to outperform
traditional signal analysis based audio codecs. This is mostly achieved by following the VQ …

In this paper, we propose an effective multi-stage neural audio coding algorithm that
encodes full-band audio signals (up to 20 kHz) using an end-to-end training criterion. By …

Most current very low bit rate (VLBR) speech coding systems use hidden Markov model
(HMM) based speech recognition and synthesis techniques. This allows transmission of …