Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.00503 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916927396380672 |
|---|---|
| author | Zuo, Jialong Zhang, Guangyan Fang, Minghui Ji, Shengpeng Jiao, Xiaoqi Li, Jingyu Guo, Yiwen Zhao, Zhou |
| author_facet | Zuo, Jialong Zhang, Guangyan Fang, Minghui Ji, Shengpeng Jiao, Xiaoqi Li, Jingyu Guo, Yiwen Zhao, Zhou |
| contents | Discrete speech representation learning has recently attracted increasing interest in both acoustic and semantic modeling. Existing approaches typically encode 16 kHz waveforms into discrete tokens at a rate of 25 or 50 tokens per second. However, given that speech generally conveys only 2 to 5 words per second, such fine-grained tokenization introduces redundancy and hinders efficiency in downstream training and inference. Moreover, semantic speech representations at this frequency primarily capture phonetic-level information, while semantic understanding may not require such detailed token-level resolution. To address these limitations, we propose an entropy-based dynamic aggregation framework for learning compressed semantic speech representations. A speech language model is first pre-trained via next-token prediction on large-scale unlabeled data to capture frequent token patterns. Predictive entropy is then used to adaptively determine aggregation boundaries, followed by a cross-attention module that fuses information within each segment. By adjusting the entropy threshold, the granularity and compression ratio of the representations can be flexibly controlled. Experiments on ASR, speech-to-text translation, and voice conversion tasks demonstrate that the compressed representations perform on par with or better than dense token sequences, demonstrating the effectiveness of the proposed approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_00503 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Entropy-based Coarse and Compressed Semantic Speech Representation Learning Zuo, Jialong Zhang, Guangyan Fang, Minghui Ji, Shengpeng Jiao, Xiaoqi Li, Jingyu Guo, Yiwen Zhao, Zhou Computation and Language Audio and Speech Processing Discrete speech representation learning has recently attracted increasing interest in both acoustic and semantic modeling. Existing approaches typically encode 16 kHz waveforms into discrete tokens at a rate of 25 or 50 tokens per second. However, given that speech generally conveys only 2 to 5 words per second, such fine-grained tokenization introduces redundancy and hinders efficiency in downstream training and inference. Moreover, semantic speech representations at this frequency primarily capture phonetic-level information, while semantic understanding may not require such detailed token-level resolution. To address these limitations, we propose an entropy-based dynamic aggregation framework for learning compressed semantic speech representations. A speech language model is first pre-trained via next-token prediction on large-scale unlabeled data to capture frequent token patterns. Predictive entropy is then used to adaptively determine aggregation boundaries, followed by a cross-attention module that fuses information within each segment. By adjusting the entropy threshold, the granularity and compression ratio of the representations can be flexibly controlled. Experiments on ASR, speech-to-text translation, and voice conversion tasks demonstrate that the compressed representations perform on par with or better than dense token sequences, demonstrating the effectiveness of the proposed approach. |
| title | Entropy-based Coarse and Compressed Semantic Speech Representation Learning |
| topic | Computation and Language Audio and Speech Processing |
| url | https://arxiv.org/abs/2509.00503 |