Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2406.08196 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913387953258496 |
|---|---|
| author | Lv, Yuanjun Li, Hai Yan, Ying Liu, Junhui Xie, Danming Xie, Lei |
| author_facet | Lv, Yuanjun Li, Hai Yan, Ying Liu, Junhui Xie, Danming Xie, Lei |
| contents | Vocoders reconstruct speech waveforms from acoustic features and play a pivotal role in modern TTS systems. Frequent-domain GAN vocoders like Vocos and APNet2 have recently seen rapid advancements, outperforming time-domain models in inference speed while achieving comparable audio quality. However, these frequency-domain vocoders suffer from large parameter sizes, thus introducing extra memory burden. Inspired by PriorGrad and SpecGrad, we employ pseudo-inverse to estimate the amplitude spectrum as the initialization roughly. This simple initialization significantly mitigates the parameter demand for vocoder. Based on APNet2 and our streamlined Amplitude prediction branch, we propose our FreeV, compared with its counterpart APNet2, our FreeV achieves 1.8 times inference speed improvement with nearly half parameters. Meanwhile, our FreeV outperforms APNet2 in resynthesis quality, marking a step forward in pursuing real-time, high-fidelity speech synthesis. Code and checkpoints is available at: https://github.com/BakerBunker/FreeV |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_08196 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FreeV: Free Lunch For Vocoders Through Pseudo Inversed Mel Filter Lv, Yuanjun Li, Hai Yan, Ying Liu, Junhui Xie, Danming Xie, Lei Sound Audio and Speech Processing Vocoders reconstruct speech waveforms from acoustic features and play a pivotal role in modern TTS systems. Frequent-domain GAN vocoders like Vocos and APNet2 have recently seen rapid advancements, outperforming time-domain models in inference speed while achieving comparable audio quality. However, these frequency-domain vocoders suffer from large parameter sizes, thus introducing extra memory burden. Inspired by PriorGrad and SpecGrad, we employ pseudo-inverse to estimate the amplitude spectrum as the initialization roughly. This simple initialization significantly mitigates the parameter demand for vocoder. Based on APNet2 and our streamlined Amplitude prediction branch, we propose our FreeV, compared with its counterpart APNet2, our FreeV achieves 1.8 times inference speed improvement with nearly half parameters. Meanwhile, our FreeV outperforms APNet2 in resynthesis quality, marking a step forward in pursuing real-time, high-fidelity speech synthesis. Code and checkpoints is available at: https://github.com/BakerBunker/FreeV |
| title | FreeV: Free Lunch For Vocoders Through Pseudo Inversed Mel Filter |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2406.08196 |