paper-with-me

홈 › Papers

Analysing the Language of Neural Audio Codecs

2025-09-01 · Joonyong Park, Shinnosuke Takamichi, David M. Chan, Shunsuke Kando, Yuki Saito, Hiroshi Saruwatari arxiv

This study presents a comparative analysis of the statistical and linguistic properties of neural audio codecs (NACs). We investigate discrete speech tokens produced by various NAC models, examining their adherence to linguistic statistical laws such as Zipf's law and Heaps' law, as well as their entropy and redundancy. To assess how these token-level properties relate to semantic and acoustic preservation in synthesized speech, we evaluate intelligibility using error rates of automatic speech recognition, and quality using the UTMOS score. Our results reveal that NAC tokens, particularly 3-grams, exhibit language-like statistical patterns. Moreover, these properties, together with measures of information content, are found to correlate with improved performances in speech recognition and resynthesis tasks. These findings offer insights into the structure of NAC token sequences and inform the design of more effective generative speech models.

📄 PDF Abstract BibTeX arXiv:2509.01390

Code (0)

등록된 구현이 없습니다.

Tasks

Speech Recognition

Similar Papers 제목 키워드 기반

Towards audio language modeling - an overview

2024-02-20 · Haibin Wu, Xuanjun Chen, Yi-Cheng Lin, Kai-Wei Chang 외

Neural audio codecs are initially introduced to compress audio data into compact codes to reduce transmission latency. Researchers recently discovered the potential of codecs as suitable tokenizers for converting continu…

Language ModelingLanguage Modelling

NanoCodec: Towards High-Quality Ultra Fast Speech LLM Inference

2025-08-07 · Edresson Casanova, Paarth Neekhara, Ryan Langman, Shehzeen Hussain 외 arxiv

Large Language Models (LLMs) have significantly advanced audio processing by leveraging audio codecs to discretize audio into tokens, enabling the application of language modeling techniques to speech data. However, exis…

STACodec: Semantic Token Assignment for Balancing Acoustic Fidelity and Semantic Information in Audio Codecs

2026-02-05 · Kaiyuan Zhang, Mohan Shi, Eray Eren, Natarajan Balaji Shankar 외 arxiv

Neural audio codecs are widely used for audio compression and can be integrated into token-based language models. Traditional codecs preserve acoustic details well but lack semantic information. Recent hybrid codecs atte…

Self-Supervised Learning

Spectral Codecs: Improving Non-Autoregressive Speech Synthesis with Spectrogram-Based Audio Codecs

2024-06-07 · Ryan Langman, Ante Jukić, Kunal Dhawan, Nithin Rao Koluguri 외

Historically, most speech models in machine-learning have used the mel-spectrogram as a speech representation. Recently, discrete audio tokens produced by neural audio codecs have become a popular alternate speech repres…

QuantizationSpeech Synthesistext-to-speechText to Speech

ERVQ: Enhanced Residual Vector Quantization with Intra-and-Inter-Codebook Optimization for Neural Audio Codecs

2024-10-16 · Rui-Chen Zheng, Hui-Peng Du, Xiao-Hang Jiang, Yang Ai 외

Current neural audio codecs typically use residual vector quantization (RVQ) to discretize speech signals. However, they often experience codebook collapse, which reduces the effective codebook size and leads to suboptim…

DiversityOnline ClusteringQuantizationtext-to-speech+1