Attentron: Few-Shot Text-to-Speech Utilizing Attention-Based Variable-Length Embedding
This addresses the need for personalized voice cloning with minimal data, offering a practical solution for applications like virtual assistants, though it appears incremental by building on existing TTS methods.
The authors tackled the problem of few-shot text-to-speech synthesis for unseen speakers by proposing Attentron, which uses attention-based variable-length embeddings and dual encoders to improve stability and quality, achieving significant outperformance over state-of-the-art models in speaker similarity and quality based on human evaluation.
On account of growing demands for personalization, the need for a so-called few-shot TTS system that clones speakers with only a few data is emerging. To address this issue, we propose Attentron, a few-shot TTS model that clones voices of speakers unseen during training. It introduces two special encoders, each serving different purposes. A fine-grained encoder extracts variable-length style information via an attention mechanism, and a coarse-grained encoder greatly stabilizes the speech synthesis, circumventing unintelligible gibberish even for synthesizing speech of unseen speakers. In addition, the model can scale out to an arbitrary number of reference audios to improve the quality of the synthesized speech. According to our experiments, including a human evaluation, the proposed model significantly outperforms state-of-the-art models when generating speech for unseen speakers in terms of speaker similarity and quality.