# Neural Feature Predictor and Discriminative Residual Coding for Low-Bitrate Speech Coding

arXiv: https://arxiv.org/abs/2211.02506 · HTML: https://mlsanity.com/paper/2211.02506/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Nov 2022.

## Built on by (most cited first, 4 of 4)

- [FunCodec: A Fundamental, Reproducible and Integrable Open-source Toolkit for Neural Speech Codec](https://mlsanity.com/paper/2309.07405/) (2023)
- [Generative De-Quantization for Neural Speech Codec via Latent Diffusion](https://mlsanity.com/paper/2311.08330/) (2023)
- [Neural Speech and Audio Coding: Modern AI Technology Meets Traditional Codecs](https://mlsanity.com/paper/2408.06954/) (2024)
- [The Equalizer: Introducing Shape-Gain Decomposition in Neural Audio Codecs](https://mlsanity.com/paper/2602.15491/) (2026)

## Abstract (arXiv)

Low and ultra-low-bitrate neural speech coding achieves unprecedented coding gain by generating speech signals from compact speech features. This paper introduces additional coding efficiency in neural speech coding by reducing the temporal redundancy existing in the frame-level feature sequence via a recurrent neural predictor. The prediction can achieve a low-entropy residual representation, which we discriminatively code based on their contribution to the signal reconstruction. The harmonization of feature prediction and discriminative coding results in a dynamic bit allocation algorithm that spends more bits on unpredictable but rare events. As a result, we develop a scalable, lightweight, low-latency, and low-bitrate neural speech coding system. We demonstrate the advantage of the proposed methods using the LPCNet as a neural vocoder. While the proposed method guarantees causality in its prediction, the subjective tests and feature space analysis show that our model achieves superior coding efficiency compared to LPCNet and Lyra V2 in the very low bitrates.