Improving Sequence-to-Sequence Acoustic Modeling by Adding Text-Supervision

Abstract

This paper presents methods of making using of text supervision to improve the performance of sequence-to-sequence (seq2seq) voice conversion. Compared with conventional frame-to-frame voice conversion approaches, the seq2seq acoustic modeling method proposed in our previous work achieved higher naturalness and similarity. In this paper, we further improve its performance by utilizing the text transcriptions of parallel training data. First, a multi-task learning structure is designed which adds auxiliary classifiers to the middle layers of the seq2seq model and predicts linguistic labels as a secondary task. Second, a data-augmentation method is proposed which utilizes text alignment to produce extra parallel sequences for model training. Experiments are conducted to evaluate our proposed method with training sets at different sizes. Experimental results show that the multi-task learning with linguistic labels is effective at reducing the errors of seq2seq voice conversion. The data-augmentation method can further improve the performance of seq2seq voice conversion when only 50 or 100 training utterances are available.

arXiv comment5 pages, 4 figures, 2 tables. Submitted to IEEE ICASSP 2019

References in corpus (5)

• Adam: A Method for Stochastic Optimization
• Neural Machine Translation by Jointly Learning to Align and Translate
• Sequence to Sequence Learning with Neural Networks
• WaveNet: A Generative Model for Raw Audio
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning

Cited by in corpus (9)

• Emotion Intensity and its Control for Emotional Voice Conversion
• Voice Transformer Network: Sequence-to-Sequence Voice Conversion Using Transformer with Text-to-Speech Pretraining
• An Overview of Voice Conversion and its Challenges: From Statistical Modeling to Deep Learning
• Mel-spectrogram augmentation for sequence to sequence voice conversion
• SpeechNet: A Universal Modularized Model for Speech Processing Tasks
• Hierarchical Sequence to Sequence Voice Conversion with Limited Data
• Limited Data Emotional Voice Conversion Leveraging Text-to-Speech: Two-stage Sequence-to-Sequence Training
• Non-autoregressive sequence-to-sequence voice conversion
• A Preliminary Study of a Two-Stage Paradigm for Preserving Speaker Identity in Dysarthric Voice Conversion