The Thinkit System for Icassp2021 M2voc Challenge
- 6 June 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 8593-8597
Abstract
In this paper, we introduce the low resource text-to-speech system from the ThinkIT team submitted to Multi-Speaker Multi-Style Voice Cloning Challenge (M2VoC). The challenge has two tasks: few-shot track1 provides 100 samples for each person and one-shot track2 offers 5 samples only. Each track contains two sub-tracks A and B. Instead of sub-track A, sub-track B can use extra public data besides the released data. But we participate in the sub-track A only. We choose the finetune as our backbone strategy. Our submitted systems include BERT based prosody boundary prediction module, FastSpeech based acoustic model to generate acoustic features from text input, and HIFIGAN based vocoder to generate waveform from acoustic features. Among them, acoustic models are susceptible to low resource speakers. To prevent over-fitting, we modified the acoustic model and split out validation set to assist the manual model selection. Evaluation results provided by the challenges organizers demonstrate the effectiveness of our system.Keywords
This publication has 8 references indexed in Scilit:
- NAUTILUS: A Versatile Voice Cloning SystemIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020
- Speaker Conditional WaveRNN: Towards Universal Neural Vocoder for Unseen Speaker and Recording Conditions2020
- Attentron: Few-Shot Text-to-Speech Utilizing Attention-Based Variable-Length Embedding2020
- BOFFIN TTS: Few-Shot Speaker Adaptation by Bayesian Optimization2020
- Zero-Shot Multi-Speaker Text-To-Speech with State-Of-The-Art Neural Speaker Embeddings2020
- Self-attention Based Prosodic Boundary Prediction for Chinese Speech Synthesis2019
- Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions2018
- Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi2017