Guided-TTS: A Diffusion Model for Text-to-Speech via Classifier Guidance

TL;DR

Guided-TTS introduces a novel diffusion-based TTS model that synthesizes speech without transcripts by leveraging classifier guidance, achieving high quality comparable to state-of-the-art models.

Contribution

It combines an unconditional diffusion model with a phoneme classifier trained on large-scale data, enabling transcript-free high-quality speech synthesis.

Findings

Achieves performance comparable to Grad-TTS without transcripts.

Performs well on diverse and long-form untranscribed datasets.

Reduces pronunciation errors with a norm-based scaling method.

Abstract

We propose Guided-TTS, a high-quality text-to-speech (TTS) model that does not require any transcript of target speaker using classifier guidance. Guided-TTS combines an unconditional diffusion probabilistic model with a separately trained phoneme classifier for classifier guidance. Our unconditional diffusion model learns to generate speech without any context from untranscribed speech data. For TTS synthesis, we guide the generative process of the diffusion model with a phoneme classifier trained on a large-scale speech recognition dataset. We present a norm-based scaling method that reduces the pronunciation errors of classifier guidance in Guided-TTS. We show that Guided-TTS achieves a performance comparable to that of the state-of-the-art TTS model, Grad-TTS, without any transcript for LJSpeech. We further demonstrate that Guided-TTS performs well on diverse datasets including a long-form untranscribed dataset.