arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.18541cs.SDcs.AIeess.AS
收藏

利用自监督语音模型提升NAM到语音合成的可懂度

Towards Improving NAM-to-Speech Synthesis Intelligibility using Self-Supervised Speech Models

IIIT Hyderabad(印度信息技术与通信技术研究院海得拉巴分校);TCS Research(塔塔咨询服务研究中心)

Neil Shah, Shirish Karande, Vineet Gandhi

AI总结:

本文提出一种利用自监督和Seq2Seq学习模拟真实语音的新方法,显著提升了NAM到语音转换的可懂度,在MCD指标上超越SOTA 29.08%,并以42.57%的WER为扩充后的语料库设定了基准。

摘要:

我们提出了一种新方法,利用自监督和序列到序列(Seq2Seq)学习技术,显著提升非可听呢喃(NAM)到语音转换任务中的可懂度。与显式记录真实语音的传统方法不同,我们的方法依赖自监督和语音到语音合成来模拟真实语音。尽管使用了模拟语音,我们的方法在Mel倒谱失真(MCD)指标上仍比当前最先进(SOTA)水平提升了29.08%。此外,我们给出了错误率,并证明了模型合成目标新声音的能力。同时,我们提出了一种扩充现有CSTR NAM TIMIT Plus语料库的方法,以42.57%的词错率(WER)为基准来衡量合成语音的可懂度。语音样本见https://nam2speech.github.io/NAM2Speech/

英文摘要:

We propose a novel approach to significantly improve the intelligibility in the Non-Audible Murmur (NAM)-to-speech conversion task, leveraging self-supervision and sequence-to-sequence (Seq2Seq) learning techniques. Unlike conventional methods that explicitly record ground-truth speech, our methodology relies on self-supervision and speech-to-speech synthesis to simulate ground-truth speech. Despite utilizing simulated speech, our method surpasses the current state-of-the-art (SOTA) by 29.08% improvement in the Mel-Cepstral Distortion (MCD) metric. Additionally, we present error rates and demonstrate our model's proficiency to synthesize speech in novel voices of interest. Moreover, we present a methodology for augmenting the existing CSTR NAM TIMIT Plus corpus, setting a benchmark with a Word Error Rate (WER) of 42.57% to gauge the intelligibility of the synthesized speech. Speech samples can be found at https://nam2speech.github.io/NAM2Speech/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑