paper-with-me

홈 › Papers

Using RLHF to align speech enhancement approaches to mean-opinion quality scores

2024-10-17 · Anurag Kumar, Andrew Perrault, Donald S. Williamson

Objective speech quality measures are typically used to assess speech enhancement algorithms, but it has been shown that they are sub-optimal as learning objectives because they do not always align well with human subjective ratings. This misalignment often results in noticeable distortions and artifacts that cause speech enhancement to be ineffective. To address these issues, we propose a reinforcement learning from human feedback (RLHF) framework to fine-tune an existing speech enhancement approach by optimizing performance using a mean-opinion score (MOS)-based reward model. Our results show that the RLHF-finetuned model has the best performance across different benchmarks for both objective and MOS-based speech quality assessment metrics on the Voicebank+DEMAND dataset. Through ablation studies, we show that both policy gradient loss and supervised MSE loss are important for balanced optimization across the different metrics.

📄 PDF Abstract BibTeX arXiv:2410.13182

Code (0)

등록된 구현이 없습니다.

Tasks

Speech Enhancement

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

DLPO: Diffusion Model Loss-Guided Reinforcement Learning for Fine-Tuning Text-to-Speech Diffusion Models

2024-05-23 · Jingyi Chen, Ju-Seung Byun, Micha Elsner, Andrew Perrault

Recent advancements in generative models have sparked a significant interest within the machine learning community. Particularly, diffusion models have demonstrated remarkable capabilities in synthesizing images and spee…

Image Generationreinforcement-learningReinforcement LearningSpeech Synthesis+3

LIRE: listwise reward enhancement for preference alignment

2024-05-22 · Mingye Zhu, Yi Liu, Lei Zhang, Junbo Guo 외

Recently, tremendous strides have been made to align the generation of Large Language Models (LLMs) with human values to mitigate toxic or unhelpful content. Leveraging Reinforcement Learning from Human Feedback (RLHF) p…

Deep learning for minimum mean-square error approaches to speech enhancement

2019-08-01 · Speech communication 2019 8 · Aaron Nicolson, Kuldip K. Paliwal

Recently, the focus of speech enhancement research has shifted from minimum mean-square error (MMSE) approaches, like the MMSE short-time spectral amplitude (MMSE-STSA) estimator, to state-of-the-art masking- and mapping…

Deep LearningSpeech Enhancement

Progressively Label Enhancement for Large Language Model Alignment

2024-08-05 · Biao Liu, Ning Xu, Xin Geng

Large Language Models (LLM) alignment aims to prevent models from producing content that misaligns with human expectations, which can lead to ethical and legal concerns. In the last few years, Reinforcement Learning from…

Language ModelingLanguage ModellingLarge Language Modelmodel

Beyond U-Net: A Latent-Representation-Aligned Skip-Free Backbone for Flow-Matching Speech Enhancement

2026-06-23 · Wangyi Pu, Michele Scarpiniti arxiv

Generative models, particularly diffusion and score-based approaches, have recently achieved strong performance in speech enhancement, but their iterative sampling process limits real-time deployment. Flow Matching offer…

Speech Enhancement