WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling

Title: WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling
Authors: Shengpeng Ji, Ziyue Jiang, Xize Cheng, Yifu Chen, Minghui Fang, Jialong Zuo, Qian Yang, Ruiqi Li, Ziang Zhang, Xiaoda Yang, Rongjie Huang, Yidi Jiang, Qian Chen, Siqi Zheng, Wen Wang, Zhou Zhao
Published: 29th August 2024 (Thursday) @ 13:43:36
Link: http://arxiv.org/abs/2408.16532v1

Abstract

Language models have been effectively applied to modeling natural signals, such as images, video, speech, and audio. A crucial component of these models is the codec tokenizer, which compresses high-dimensional natural signals into lower-dimensional discrete tokens. In this paper, we introduce WavTokenizer, which offers several advantages over previous SOTA acoustic codec models in the audio domain: 1)extreme compression. By compressing the layers of quantizers and the temporal dimension of the discrete codec, one-second audio of 24kHz sampling rate requires only a single quantizer with 40 or 75 tokens. 2)improved subjective quality. Despite the reduced number of tokens, WavTokenizer achieves state-of-the-art reconstruction quality with outstanding UTMOS scores and inherently contains richer semantic information. Specifically, we achieve these results by designing a broader VQ space, extended contextual windows, and improved attention networks, as well as introducing a powerful multi-scale discriminator and an inverse Fourier transform structure. We conducted extensive reconstruction experiments in the domains of speech, audio, and music. WavTokenizer exhibited strong performance across various objective and subjective metrics compared to state-of-the-art models. We also tested semantic information, VQ utilization, and adaptability to generative models. Comprehensive ablation studies confirm the necessity of each module in WavTokenizer. The related code, demos, and pre-trained models are available at https://github.com/jishengpeng/WavTokenizer.

---

🏋️‍♀️ Checkpoints available on HF and include:

• novateur/WavTokenizer-medium-speech-75token (Updated Sep 19)
• novateur/WavTokenizer-medium-music-audio-75token (Updated Sep 19)
• novateur/WavTokenizer-large-unify-75token (Updated Sep 9)
• novateur/WavTokenizer-large-unify-40token (Updated Oct 22)
• novateur/WavTokenizer-large-speech-75token (Updated Oct 22)

Code: https://github.com/jishengpeng/WavTokenizer

From the repo README:

• 2024.10.22: We update WavTokenizer on arxiv and release WavTokenizer-Large checkpoint.
• 2024.09.09: We release WavTokenizer-medium checkpoint on huggingface

---

Advantages over previous state-of-the-art (SOTA) acoustic codec models in the audio domain:

1. Extreme Compression. By compressing the layers of quantizers and the temporal dimension of the discrete codec, one-second audio of 24kHz sampling rate requires only a single quantizer with 40 or 75 tokens.
2. Improved Subjective Reconstruction Quality. Despite the reduced number of tokens, WavTokenizer achieves SOTA reconstruction quality with outstanding UTMOS scores and also inherently contains richer semantic information

Achieve these results by:

• designing a broader VQ space
• extending contextual windows
• improving attention networks
• introducing a powerful multi-scale discriminator and an inverse Fourier transform structure