FreeV: Free Lunch For Vocoders Through Pseudo Inversed Mel Filter

Yuanjun Lv, Hai Li, Ying Yan, Junhui Liu, Danming Xie, Lei Xie

科研成果: 期刊稿件会议文章同行评审

摘要

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× 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.

源语言英语
页(从-至)3869-3873
页数5
期刊Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
DOI
出版状态已出版 - 2024
活动25th Interspeech Conferece 2024 - Kos Island, 希腊
期限: 1 9月 20245 9月 2024

指纹

探究 'FreeV: Free Lunch For Vocoders Through Pseudo Inversed Mel Filter' 的科研主题。它们共同构成独一无二的指纹。

引用此