TextrolSpeech: A Text Style Control Speech Corpus With Codec Language Text-to-Speech Models
This work addresses the problem of controllable text-to-speech for researchers and developers by providing a novel dataset and model, though it appears incremental as it builds on existing TTS and language model approaches.
The paper tackles the challenge of generating speech from natural text prompts by introducing TextrolSpeech, a large-scale dataset with 236,220 pairs of text descriptions and speech samples annotated with five style factors, and proposes Salle, an efficient architecture that treats text-controllable TTS as a language model task using audio codec codes, achieving comparable performance in controllable TTS tasks.
Recently, there has been a growing interest in the field of controllable Text-to-Speech (TTS). While previous studies have relied on users providing specific style factor values based on acoustic knowledge or selecting reference speeches that meet certain requirements, generating speech solely from natural text prompts has emerged as a new challenge for researchers. This challenge arises due to the scarcity of high-quality speech datasets with natural text style prompt and the absence of advanced text-controllable TTS models. In light of this, 1) we propose TextrolSpeech, which is the first large-scale speech emotion dataset annotated with rich text attributes. The dataset comprises 236,220 pairs of style prompt in natural text descriptions with five style factors and corresponding speech samples. Through iterative experimentation, we introduce a multi-stage prompt programming approach that effectively utilizes the GPT model for generating natural style descriptions in large volumes. 2) Furthermore, to address the need for generating audio with greater style diversity, we propose an efficient architecture called Salle. This architecture treats text controllable TTS as a language model task, utilizing audio codec codes as an intermediate representation to replace the conventional mel-spectrogram. Finally, we successfully demonstrate the ability of the proposed model by showing a comparable performance in the controllable TTS task. Audio samples are available at https://sall-e.github.io/