Tara N. Sainath, Brian Kingsbury, Abdel-rahman Mohamed, B. Ramabhadran
{"title":"在深度神经网络框架内学习滤波器组","authors":"Tara N. Sainath, Brian Kingsbury, Abdel-rahman Mohamed, B. Ramabhadran","doi":"10.1109/ASRU.2013.6707746","DOIUrl":null,"url":null,"abstract":"Mel-filter banks are commonly used in speech recognition, as they are motivated from theory related to speech production and perception. While features derived from mel-filter banks are quite popular, we argue that this filter bank is not really an appropriate choice as it is not learned for the objective at hand, i.e. speech recognition. In this paper, we explore replacing the filter bank with a filter bank layer that is learned jointly with the rest of a deep neural network. Thus, the filter bank is learned to minimize cross-entropy, which is more closely tied to the speech recognition objective. On a 50-hour English Broadcast News task, we show that we can achieve a 5% relative improvement in word error rate (WER) using the filter bank learning approach, compared to having a fixed set of filters.","PeriodicalId":265258,"journal":{"name":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"170","resultStr":"{\"title\":\"Learning filter banks within a deep neural network framework\",\"authors\":\"Tara N. Sainath, Brian Kingsbury, Abdel-rahman Mohamed, B. Ramabhadran\",\"doi\":\"10.1109/ASRU.2013.6707746\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Mel-filter banks are commonly used in speech recognition, as they are motivated from theory related to speech production and perception. While features derived from mel-filter banks are quite popular, we argue that this filter bank is not really an appropriate choice as it is not learned for the objective at hand, i.e. speech recognition. In this paper, we explore replacing the filter bank with a filter bank layer that is learned jointly with the rest of a deep neural network. Thus, the filter bank is learned to minimize cross-entropy, which is more closely tied to the speech recognition objective. On a 50-hour English Broadcast News task, we show that we can achieve a 5% relative improvement in word error rate (WER) using the filter bank learning approach, compared to having a fixed set of filters.\",\"PeriodicalId\":265258,\"journal\":{\"name\":\"2013 IEEE Workshop on Automatic Speech Recognition and Understanding\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"170\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2013 IEEE Workshop on Automatic Speech Recognition and Understanding\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ASRU.2013.6707746\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 IEEE Workshop on Automatic Speech Recognition and Understanding","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ASRU.2013.6707746","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Learning filter banks within a deep neural network framework
Mel-filter banks are commonly used in speech recognition, as they are motivated from theory related to speech production and perception. While features derived from mel-filter banks are quite popular, we argue that this filter bank is not really an appropriate choice as it is not learned for the objective at hand, i.e. speech recognition. In this paper, we explore replacing the filter bank with a filter bank layer that is learned jointly with the rest of a deep neural network. Thus, the filter bank is learned to minimize cross-entropy, which is more closely tied to the speech recognition objective. On a 50-hour English Broadcast News task, we show that we can achieve a 5% relative improvement in word error rate (WER) using the filter bank learning approach, compared to having a fixed set of filters.