CASIA OpenIR  > 复杂系统认知与决策实验室  > 听觉模型与认知计算
Lead ASR Models to Generalize Better Using Approximated Bias-Variance Tradeof
Wang FY(王方圆)1; Ming Hao2; Yuhai Shi2; Bo Xu1
2023
Conference NameICONIP 2023
Conference Date2023.11.13
Conference Placechangsha,China
Country日本
Abstract

The conventional recipe for Automatic Speech Recognition (ASR) models is to 1) train multiple checkpoints on a training set while relying on a validation set to prevent over fitting using early stopping and 2) average several last checkpoints or that of the lowest validation losses to obtain the final model. In this paper, we rethink and update the early stopping and checkpoint averaging from the perspective of the bias-variance tradeoff. Theoretically, the bias and variance represent the fitness and variability of a model and the tradeoff of them determines the overall generalization error. But, it’s impractical to evaluate them precisely. As an alternative, we take the training loss and validation loss as proxies of bias and variance and guide the early stopping and checkpoint averaging using their tradeoff, namely an Approximated Bias-Variance Tradeoff  ApproBiVT). When evaluating with advanced ASR models, our recipe provides 2.5%–3.7% and 3.1%–4.6% CER reduction on the AISHELL-1 and AISHELL-2, respectively (The code and sampled unaugmented training sets used in this paper will be public available on GitHub).

Sub direction classification语音识别与合成
planning direction of the national heavy laboratory语音语言处理
Paper associated data
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/57379
Collection复杂系统认知与决策实验室_听觉模型与认知计算
Affiliation1.中国科学院自动化研究所
2.广播科学院互联网所
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Wang FY,Ming Hao,Yuhai Shi,et al. Lead ASR Models to Generalize Better Using Approximated Bias-Variance Tradeof[C],2023.
Files in This Item: Download All
File Name/Size DocType Version Access License
iconip2023_published(1933KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang FY(王方圆)]'s Articles
[Ming Hao]'s Articles
[Yuhai Shi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang FY(王方圆)]'s Articles
[Ming Hao]'s Articles
[Yuhai Shi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang FY(王方圆)]'s Articles
[Ming Hao]'s Articles
[Yuhai Shi]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: iconip2023_published.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.