Effect of Data Reduction on Sequence-to-sequence Neural TTS
Authors:
- 1 May 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 7075-7079
Abstract
Recent speech synthesis systems based on sampling from autoregressive neural network models can generate speech almost indistinguishable from human recordings. However, these models require large amounts of data. This paper shows that the lack of data from one speaker can be compensated with data from other speakers. The naturalness of Tacotron2-like models trained on a blend of 5k utterances from 7 speakers is better than or equivalent to that of speaker dependent models trained on 15k utterances. Additionally, in terms of stability multi-speaker models are always more stable. We also demonstrate that models mixing only 1250 utterances from a target speaker with 5k utterances from another 6 speakers can produce significantly better quality than state-of-the-art DNN-guided unit selection systems trained on more than 10 times the data from the target speaker.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Comprehensive Evaluation of Statistical Speech Waveform Synthesis2018
- Experiments with Training Corpora for Statistical Text-to-speech Systems.2018
- Investigating different representations for modeling and controlling multiple emotions in DNN-based speech synthesisSpeech Communication, 2018
- Tacotron: Towards End-to-End Speech Synthesis2017
- librosa: Audio and Music Signal Analysis in Python2015
- Measuring a decade of progress in Text-to-SpeechLoquens, 2014
- Analysis of Speaker Adaptation Algorithms for HMM-Based Speech Synthesis and a Constrained SMAPLR Adaptation AlgorithmIEEE Transactions on Audio, Speech, and Language Processing, 2009