CM-TTS: Enhancing Real Time Text-to-Speech Synthesis Efficiency through Weighted Samplers and Consistency Models

Xiang Li, FanBu FanBu, Ambuj Mehrish, Yingting Li, Jiale Han, Bo Cheng, Soujanya Poria


Abstract
Neural Text-to-Speech (TTS) systems find broad applications in voice assistants, e-learning, and audiobook creation. The pursuit of modern models, like Diffusion Models (DMs), holds promise for achieving high-fidelity, real-time speech synthesis. Yet, the efficiency of multi-step sampling in Diffusion Models presents challenges. Efforts have been made to integrate GANs with DMs, speeding up inference by approximating denoising distributions, but this introduces issues with model convergence due to adversarial training. To overcome this, we introduce CM-TTS, a novel architecture grounded in consistency models (CMs). Drawing inspiration from continuous-time diffusion models, CM-TTS achieves top-quality speech synthesis in fewer steps without adversarial training or pre-trained model dependencies. We further design weighted samplers to incorporate different sampling positions into model training with dynamic probabilities, ensuring unbiased learning throughout the entire training process. We present a real-time mel-spectrogram generation consistency model, validated through comprehensive evaluations. Experimental results underscore CM-TTS’s superiority over existing single-step speech synthesis systems, representing a significant advancement in the field.
Anthology ID:
2024.findings-naacl.240
Volume:
Findings of the Association for Computational Linguistics: NAACL 2024
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Kevin Duh, Helena Gomez, Steven Bethard
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3777–3794
Language:
URL:
https://aclanthology.org/2024.findings-naacl.240
DOI:
10.18653/v1/2024.findings-naacl.240
Bibkey:
Cite (ACL):
Xiang Li, FanBu FanBu, Ambuj Mehrish, Yingting Li, Jiale Han, Bo Cheng, and Soujanya Poria. 2024. CM-TTS: Enhancing Real Time Text-to-Speech Synthesis Efficiency through Weighted Samplers and Consistency Models. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 3777–3794, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
CM-TTS: Enhancing Real Time Text-to-Speech Synthesis Efficiency through Weighted Samplers and Consistency Models (Li et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/2024.findings-naacl.240.pdf