Cross-Lingual Voice Conversion with Non-Parallel Data

TitleCross-Lingual Voice Conversion with Non-Parallel Data
Publication TypeMaster Thesis
Year of Publication2017
AuthorsAlonso-Jiménez, P.
AbstractIn this project a Phonetic Posteriorgram (PPG) based Voice Conversion system is implemented. The main goal is to perform and evaluate conversions of singing voice. The cross-gender and cross-lingual scenarios are considered. Additionally, the use of spectral envelope based MFCC and pseudo-singing dataset for ASR training are proposed in order to improve the performance of the system in the singing context.
KeywordsSI-ASR, Voice Conversion, Voice Synthesis
Final publicationhttps://doi.org/10.5281/zenodo.1117153
intranet