Parallel-Data-Free Many-to-Many Voice Conversion Based on DNN Integrated with Eigenspace Using a Non-Parallel Speech Corpus

2017 ◽  
Author(s):  
Tetsuya Hashimoto ◽  
Hidetsugu Uchida ◽  
Daisuke Saito ◽  
Nobuaki Minematsu
Keyword(s):  
Voice Conversion ◽  
Speech Corpus ◽  
Parallel Data

Average Modeling Approach to Voice Conversion with Non-Parallel Data

2018 ◽  
Author(s):  
Xiaohai Tian ◽  
Junchao Wang ◽  
Haihua Xu ◽  
Eng-Siong Chng ◽  
Haizhou Li
Keyword(s):  
Voice Conversion ◽  
Modeling Approach ◽  
Parallel Data

scholarly journals Non-parallel dictionary learning for voice conversion using non-negative Tucker decomposition

2019 ◽  
Vol 2019 (1) ◽  
Author(s):  
Yuki Takashima ◽  
Toru Nakashika ◽  
Tetsuya Takiguchi ◽  
Yasuo Ariki
Keyword(s):  
Dictionary Learning ◽  
Computational Cost ◽  
Tensor Decomposition ◽  
Gaussian Mixture ◽  
Voice Conversion ◽  
Specific Information ◽  
Learning Method ◽  
Tucker Decomposition ◽  
Parallel Data ◽  
High Computational Cost

Abstract Voice conversion (VC) is a technique of exclusively converting speaker-specific information in the source speech while preserving the associated phonemic information. Non-negative matrix factorization (NMF)-based VC has been widely researched because of the natural-sounding voice it achieves when compared with conventional Gaussian mixture model-based VC. In conventional NMF-VC, models are trained using parallel data which results in the speech data requiring elaborate pre-processing to generate parallel data. NMF-VC also tends to be an extensive model as this method has several parallel exemplars for the dictionary matrix, leading to a high computational cost. In this study, an innovative parallel dictionary-learning method using non-negative Tucker decomposition (NTD) is proposed. The proposed method uses tensor decomposition and decomposes an input observation into a set of mode matrices and one core tensor. The proposed NTD-based dictionary-learning method estimates the dictionary matrix for NMF-VC without using parallel data. The experimental results show that the proposed method outperforms other methods in both parallel and non-parallel settings.

Construction and annotation of a corpus of contemporary Nepali

Corpora ◽  
2008 ◽  
Vol 3 (2) ◽  
pp. 213-225 ◽  
Author(s):  
Yogendra P. Yadava ◽  
Andrew Hardie ◽  
Ram Raj Lohani ◽  
Bhim N. Regmi ◽  
Srishtee Gurung ◽  
...  
Keyword(s):  
Speech Corpus ◽  
Part Of Speech ◽  
Parallel Data ◽  
National Corpus

In this paper, we describe the construction of the 14-million-word Nepali National Corpus (NNC). This corpus includes both spoken and written data, the latter incorporating a Nepali match for FLOB and a broader collection of text. Additional resources within the NNC include parallel data (English–Nepali and Nepali–English) and a speech corpus. The NNC is encoded as Unicode text and marked up in CES-compatible XML. The whole corpus is also annotated with part-of-speech tags. We describe the process of devising a tagset and retraining tagger software for the Nepali language, for which there were no existing corpus resources. Finally, we explore some present and future applications of the corpus, including lexicography, NLP, and grammatical research.

scholarly journals Text-independent F0 transformation with non-parallel data for voice conversion

2010 ◽  
Author(s):  
Zhi-Zheng Wu ◽  
Tomi Kinnunen ◽  
Eng Siong Chng ◽  
Haizhou Li
Keyword(s):  
Voice Conversion ◽  
Parallel Data
Sign in / Sign up

Export Citation Format

Share Document