Yes, the style labels primarily refer to emotions, different speaking styles, and so on. In most datasets like LibriTTS and LJSpeech, the speaking style is quite consistent. However, you can search for datasets that provide a wider range of emotions and diverse speaking styles.
Yes, the style labels primarily refer to emotions, different speaking styles, and so on. In most datasets like LibriTTS and LJSpeech, the speaking style is quite consistent. However, you can search for datasets that provide a wider range of emotions and diverse speaking styles.