A Neural Text-to-Speech Model Utilizing Broadcast Data Mixed with Background Music
This addresses a domain-specific challenge for TTS systems using noisy media data, but it is incremental as it builds on existing GST-TTS frameworks.
The paper tackled the problem of training neural text-to-speech models with broadcast data that includes background music, proposing a method with a music filter and auxiliary quality classifier, which synthesized higher-quality speech than conventional methods.
Recently, it has become easier to obtain speech data from various media such as the internet or YouTube, but directly utilizing them to train a neural text-to-speech (TTS) model is difficult. The proportion of clean speech is insufficient and the remainder includes background music. Even with the global style token (GST). Therefore, we propose the following method to successfully train an end-to-end TTS model with limited broadcast data. First, the background music is removed from the speech by introducing a music filter. Second, the GST-TTS model with an auxiliary quality classifier is trained with the filtered speech and a small amount of clean speech. In particular, the quality classifier makes the embedding vector of the GST layer focus on representing the speech quality (filtered or clean) of the input speech. The experimental results verified that the proposed method synthesized much more high-quality speech than conventional methods.