RepCodec: A Speech Representation Codec for Speech Tokenization
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917728286146560 |
|---|---|
| author | Huang, Zhichao Meng, Chutong Ko, Tom |
| author_facet | Huang, Zhichao Meng, Chutong Ko, Tom |
| contents | With recent rapid growth of large language models (LLMs), discrete speech tokenization has played an important role for injecting speech into LLMs. However, this discretization gives rise to a loss of information, consequently impairing overall performance. To improve the performance of these discrete speech tokens, we present RepCodec, a novel speech representation codec for semantic speech tokenization. In contrast to audio codecs which reconstruct the raw audio, RepCodec learns a vector quantization codebook through reconstructing speech representations from speech encoders like HuBERT or data2vec. Together, the speech encoder, the codec encoder and the vector quantization codebook form a pipeline for converting speech waveforms into semantic tokens. The extensive experiments illustrate that RepCodec, by virtue of its enhanced information retention capacity, significantly outperforms the widely used k-means clustering approach in both speech understanding and generation. Furthermore, this superiority extends across various speech encoders and languages, affirming the robustness of RepCodec. We believe our method can facilitate large language modeling research on speech processing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2309_00169 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | RepCodec: A Speech Representation Codec for Speech Tokenization Huang, Zhichao Meng, Chutong Ko, Tom Audio and Speech Processing Machine Learning Sound With recent rapid growth of large language models (LLMs), discrete speech tokenization has played an important role for injecting speech into LLMs. However, this discretization gives rise to a loss of information, consequently impairing overall performance. To improve the performance of these discrete speech tokens, we present RepCodec, a novel speech representation codec for semantic speech tokenization. In contrast to audio codecs which reconstruct the raw audio, RepCodec learns a vector quantization codebook through reconstructing speech representations from speech encoders like HuBERT or data2vec. Together, the speech encoder, the codec encoder and the vector quantization codebook form a pipeline for converting speech waveforms into semantic tokens. The extensive experiments illustrate that RepCodec, by virtue of its enhanced information retention capacity, significantly outperforms the widely used k-means clustering approach in both speech understanding and generation. Furthermore, this superiority extends across various speech encoders and languages, affirming the robustness of RepCodec. We believe our method can facilitate large language modeling research on speech processing. |
| title | RepCodec: A Speech Representation Codec for Speech Tokenization |
| topic | Audio and Speech Processing Machine Learning Sound |
| url | https://arxiv.org/abs/2309.00169 |