Vector quantization in speech coding

By John I. Makhoul, Salim Roukos, Herbert Gish. 1985

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

SoundStream: An End-to-End Neural Audio Codec (2021, methodology): “Vector quantization has been commonly used as a building block of traditional audio codecs [29].”
Audiodec: An Open-Source Streaming High-Fidelity Neural Audio Codec (2023, methodology): “To jointly optimize all modules, many E2E AEs adopting vector quantization (VQ) [30] have been proposed to tackle the gradient of the VQ.”
Cascaded Cross-Module Residual Learning towards Lightweight End-to-End Speech Coding (2019, methodology): “A speech coding system can be formulated by DNN as an autoencoder (AE) with a code layer discretized by vector quantization (VQ) [9] or bitwise network techniques [10], etc.”
Sequential scalar quantization of color images (1994, unknown): “The i -th experimental MSE term in (13) is equated with the corresponding i-th theoretical MSE term in (6a) to yield empirical values αe, βe, ηe α e = N1 2d1 ; βe = N2 2 N1 2 d2 ; ηe = N(2) N2 2 d3 .”
Neural Networks for Vector Quantization of Speech and Images (1990, methodology): “T coding of speech and images, and vector quantization (VQ) has emerged in recent years as a powerful technique that can provide large reductions in bit rate while preserving the essential signal characteristics [2], [ 3 ].”
Introduction to Data Compression (2005, background): “Gish, in the November 1985 issue of the Proceedings of the IEEE [161].”
Essentials of the self-organizing map (2013, background): “For classical references, cf., e.g., Gersho (1979), Gray (1984), and Makhoul, Roucos, and Gis (1985).”