Non-autoregressive sequence-to-sequence voice conversion
This work addresses voice conversion for speech synthesis applications, but it is incremental as it adapts existing non-autoregressive methods from text-to-speech to a related domain.
The paper tackles voice conversion by proposing a non-autoregressive sequence-to-sequence model based on FastSpeech2, which achieves more stable, faster, and better conversion than autoregressive models like Tacotron2 and Transformer, as demonstrated on a Japanese dataset with 1,000 utterances.
This paper proposes a novel voice conversion (VC) method based on non-autoregressive sequence-to-sequence (NAR-S2S) models. Inspired by the great success of NAR-S2S models such as FastSpeech in text-to-speech (TTS), we extend the FastSpeech2 model for the VC problem. We introduce the convolution-augmented Transformer (Conformer) instead of the Transformer, making it possible to capture both local and global context information from the input sequence. Furthermore, we extend variance predictors to variance converters to explicitly convert the source speaker's prosody components such as pitch and energy into the target speaker. The experimental evaluation with the Japanese speaker dataset, which consists of male and female speakers of 1,000 utterances, demonstrates that the proposed model enables us to perform more stable, faster, and better conversion than autoregressive S2S (AR-S2S) models such as Tacotron2 and Transformer.