Fine Tuning and Comparing Tacotron 2, Deep Voice 3, and FastSpeech 2 TTS Models in a Low Resource Environment

Author:

Gopalakrishnan T.1,Imam Syed Ayaz2,Aggarwal Archit3

Affiliation:

1. Manipal Institute of Technology,Department of Information Technology,Bengaluru

2. Vellore Institute of Technology,School of Computer Science and Engineering,Vellore,India

3. Vellore Institute of Technology,School of Computer Science and Engineering,Manipal,India

Publisher

IEEE

Reference29 articles.

1. Char2wav: End-to-end speech synthesis;sotelo,2017

2. Towards transfer learning for end-to-end speech synthesis from deep pre-trained language models;fang;arXiv preprint arXiv 1906 07165,2019

3. Improving language understanding by generative pre-training;radford,2018

4. Deep contextualized word representations;peters;Arxiv ArXiv,1802

5. Universal Language Model Fine-tuning for Text Classification

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. A Novel Dysarthric Speech Synthesis system using Tacotron2 for specific and OOV words;2024 International Conference on Signal Processing and Communications (SPCOM);2024-07-01

2. Effects of Training Strategies and the Amount of Speech Data on the Quality of Speech Synthesis;Lecture Notes in Computer Science;2024

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3