StarGANv2-VC: A Diverse, Unsupervised, Non-parallel Framework for Natural-Sounding Voice Conversion
This provides a versatile and efficient solution for voice conversion tasks, enabling applications like cross-lingual and emotional speech synthesis without parallel data, though it is incremental as it builds on existing GAN architectures.
The authors tackled unsupervised non-parallel many-to-many voice conversion by developing StarGANv2-VC, a GAN-based framework that significantly outperforms previous models and produces natural-sounding voices close to state-of-the-art TTS-based methods without text labels.
We present an unsupervised non-parallel many-to-many voice conversion (VC) method using a generative adversarial network (GAN) called StarGAN v2. Using a combination of adversarial source classifier loss and perceptual loss, our model significantly outperforms previous VC models. Although our model is trained only with 20 English speakers, it generalizes to a variety of voice conversion tasks, such as any-to-many, cross-lingual, and singing conversion. Using a style encoder, our framework can also convert plain reading speech into stylistic speech, such as emotional and falsetto speech. Subjective and objective evaluation experiments on a non-parallel many-to-many voice conversion task revealed that our model produces natural sounding voices, close to the sound quality of state-of-the-art text-to-speech (TTS) based voice conversion methods without the need for text labels. Moreover, our model is completely convolutional and with a faster-than-real-time vocoder such as Parallel WaveGAN can perform real-time voice conversion.