paper-with-me

홈 › Papers

DiffNorm: Self-Supervised Normalization for Non-autoregressive Speech-to-speech Translation

2024-05-22 · Weiting Tan, Jingyu Zhang, Lingfeng Shen, Daniel Khashabi, Philipp Koehn

Non-autoregressive Transformers (NATs) are recently applied in direct speech-to-speech translation systems, which convert speech across different languages without intermediate text data. Although NATs generate high-quality outputs and offer faster inference than autoregressive models, they tend to produce incoherent and repetitive results due to complex data distribution (e.g., acoustic and linguistic variations in speech). In this work, we introduce DiffNorm, a diffusion-based normalization strategy that simplifies data distributions for training NAT models. After training with a self-supervised noise estimation objective, DiffNorm constructs normalized target data by denoising synthetically corrupted speech features. Additionally, we propose to regularize NATs with classifier-free guidance, improving model robustness and translation quality by randomly dropping out source information during training. Our strategies result in a notable improvement of about +7 ASR-BLEU for English-Spanish (En-Es) and +2 ASR-BLEU for English-French (En-Fr) translations on the CVSS benchmark, while attaining over 14x speedup for En-Es and 5x speedup for En-Fr translations compared to autoregressive baselines.

📄 PDF Abstract BibTeX arXiv:2405.13274

Code (1)

steventan0110/diffnorm 공식 구현 pytorch

Tasks

DenoisingNoise EstimationSpeech-to-Speech TranslationTranslation

Similar Papers 제목 키워드 기반

TokAN: Accent Normalization Using Self-Supervised Speech Tokens

2026-07-04 · Qibing Bai, Shuai Wang, Yuhan Du, Bohan Li 외 arxiv

Accent normalization (AN) seeks to convert non-native (L2) accented speech into standard (L1) speech while preserving speaker identity. The current techniques either require naturally recorded parallel L1-L2 speech for t…

Reinforcement Learning

Towards Better Disentanglement in Non-Autoregressive Zero-Shot Expressive Voice Conversion

2025-06-04 · Seymanur Aktı, Tuan Nam Nguyen, Alexander Waibel

Expressive voice conversion aims to transfer both speaker identity and expressive attributes from a target speech to a given source speech. In this work, we improve over a self-supervised, non-autoregressive framework wi…

DisentanglementStyle TransferVoice Conversion

TranSpeech: Speech-to-Speech Translation With Bilateral Perturbation

2022-05-25 · Rongjie Huang, Jinglin Liu, Huadai Liu, Yi Ren 외

Direct speech-to-speech translation (S2ST) with discrete units leverages recent progress in speech representation learning. Specifically, a sequence of discrete representations derived in a self-supervised manner are pre…

Representation LearningRhythmSpeech Representation LearningSpeech Synthesis+2

Training Autoregressive Speech Recognition Models with Limited in-domain Supervision

2022-10-27 · Chak-Fai Li, Francis Keith, William Hartmann, Matthew Snover

Advances in self-supervised learning have significantly reduced the amount of transcribed audio required for training. However, the majority of work in this area is focused on read speech. We explore limited supervision …

DecoderSelf-Supervised Learningspeech-recognitionSpeech Recognition

Non-Autoregressive Predictive Coding for Learning Speech Representations from Local Dependencies

2020-11-01 · Alexander H. Liu, Yu-An Chung, James Glass

Self-supervised speech representations have been shown to be effective in a variety of speech applications. However, existing representation learning methods generally rely on the autoregressive model and/or observed glo…

Representation Learning