paper-with-me

홈 › Papers

Comparison of Multilingual Self-Supervised and Weakly-Supervised Speech Pre-Training for Adaptation to Unseen Languages

2023-05-21 · Andrew Rouditchenko, Sameer Khurana, Samuel Thomas, Rogerio Feris, Leonid Karlinsky, Hilde Kuehne, David Harwath, Brian Kingsbury, James Glass

Recent models such as XLS-R and Whisper have made multilingual speech technologies more accessible by pre-training on audio from around 100 spoken languages each. However, there are thousands of spoken languages worldwide, and adapting to new languages is an important problem. In this work, we aim to understand which model adapts better to languages unseen during pre-training. We fine-tune both models on 13 unseen languages and 18 seen languages. Our results show that the number of hours seen per language and language family during pre-training is predictive of how the models compare, despite the significant differences in the pre-training methods.

📄 PDF Abstract BibTeX arXiv:2305.12606

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Low-Resourced Speech Recognition for Iu Mien Language via Weakly-Supervised Phoneme-based Multilingual Pre-training

2024-07-18 · Lukuan Dong, Donghong Qin, Fengbo Bai, Fanhua Song 외

The mainstream automatic speech recognition (ASR) technology usually requires hundreds to thousands of hours of annotated speech data. Three approaches to low-resourced ASR are phoneme or subword based supervised pre-tra…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

XLST: Cross-lingual Self-training to Learn Multilingual Representation for Low Resource Speech Recognition

2021-03-15 · Zi-Qiang Zhang, Yan Song, Ming-Hui Wu, Xin Fang 외

In this paper, we propose a weakly supervised multilingual representation learning framework, called cross-lingual self-training (XLST). XLST is able to utilize a small amount of annotated data from high-resource languag…

Data AugmentationRepresentation Learningspeech-recognitionSpeech Recognition

Deploying self-supervised learning in the wild for hybrid automatic speech recognition

2022-05-17 · Mostafa Karimi, Changliang Liu, Kenichi Kumatani, Yao Qian 외

Self-supervised learning (SSL) methods have proven to be very successful in automatic speech recognition (ASR). These great improvements have been reported mostly based on highly curated datasets such as LibriSpeech for …

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Event DetectionScheduling+3

Large scale weakly and semi-supervised learning for low-resource video ASR

2020-05-16 · Kritika Singh, Vimal Manohar, Alex Xiao, Sergey Edunov 외

Many semi- and weakly-supervised approaches have been investigated for overcoming the labeling cost of building high quality speech recognition systems. On the challenging task of transcribing social media videos in low-…

Decoderspeech-recognitionSpeech Recognition

Weakly and Self-Supervised Class-Agnostic Motion Prediction for Autonomous Driving

2025-09-16 · Ruibo Li, Hanyu Shi, Zhe Wang, Guosheng Lin arxiv

Understanding motion in dynamic environments is critical for autonomous driving, thereby motivating research on class-agnostic motion prediction. In this work, we investigate weakly and self-supervised class-agnostic mot…

Self-Supervised LearningAutonomous DrivingScene ParsingPoint Clouds