scholarly journals The Accented English Speech Recognition Challenge 2020: Open Datasets, Tracks, Baselines, Results and Methods

2021 ◽  
Author(s):  
Xian Shi ◽  
Fan Yu ◽  
Yizhou Lu ◽  
Yuhao Liang ◽  
Qiangze Feng ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Accented English ◽  
Open Datasets

Automatic speech recognition of multiple accented English data

2010 ◽  
Author(s):  
Dimitra Vergyri ◽  
Lori Lamel ◽  
Jean-Luc Gauvain
Keyword(s):  
Speech Recognition ◽  
Automatic Speech Recognition ◽  
Accented English

scholarly journals The SLT 2021 Children Speech Recognition Challenge: Open Datasets, Rules and Baselines

2021 ◽  
Author(s):  
Fan Yu ◽  
Zhuoyuan Yao ◽  
Xiong Wang ◽  
Keyu An ◽  
Lei Xie ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Open Datasets ◽  
Children Speech

scholarly journals Computational Approaches to Exploring Persian-Accented English

2015 ◽  
Vol 13 (1) ◽  
pp. 51-60
Author(s):  
Corey Miller
Keyword(s):  
Speech Recognition ◽  
Computational Approaches ◽  
Accented English ◽  
L2 Pronunciation

Methods involving phonetic speech recognition are discussed for detecting Persianaccented English. These methods offer promise for both the identification and mitigation of L2 pronunciation errors. Pronunciation errors, both segmental and suprasegmental, particular to Persian speakers of English are discussed.

The Adaptation to Mandarin-Accented English by Older, Hearing-Impaired Listeners Following Brief Exposure to the Accent

2020 ◽  
Vol 63 (3) ◽  
pp. 858-871
Author(s):  
Jutta A. Hau ◽  
Colleen M. Holt ◽  
Sue Finch ◽  
Richard C. Dowell
Keyword(s):  
Speech Recognition ◽  
Language Processing ◽  
Hearing Impaired ◽  
Control Group ◽  
Training Study ◽  
Accented Speech ◽  
Brief Training ◽  
Australian English ◽  
Accented English ◽  
Moderate Hearing Loss

Purpose The language processing of Mandarin-accented English (MAE) by older hearing-impaired (OHI), older normally hearing (NH), and younger NH listeners was explored. We examined whether OHI adults have more difficulty than NH listeners in recognizing and adapting to MAE speech productions after receiving brief training with the accent. Method Talker-independent adaptation was evaluated in an exposure training study design. Listeners were trained either by four MAE talkers or four Australian English talkers (control group) before listening to sentences presented by a novel MAE talker. Speech recognition for both the training sentences and the experimental sentences were compared between listener groups and between the training accents. Results Listeners in all three groups (OHI, older NH, younger NH) who had been trained by the MAE talkers showed higher odds of speech recognition than listeners trained by the Australian English talkers. The OHI listeners adapted to MAE to the same degree as the NH groups despite returning lower overall odds of recognizing MAE speech. Conclusions Older listeners with mild-to-moderate hearing loss were able to benefit as much from brief exposure to MAE as did the NH groups. This encouraging result suggests that OHI listeners have access to and can exploit the information present in a relatively brief sample of accented speech and generalize their learning to a novel MAE talker.

scholarly journals Speech-on-speech Masking with Variable Access to the Linguistic Content of the Masker Speech for Native and Nonnative English Speakers

2014 ◽  
Vol 25 (04) ◽  
pp. 355-366 ◽  
Author(s):  
Lauren Calandruccio ◽  
Ann R. Bradlow ◽  
Sumitrajit Dhar
Keyword(s):  
Hearing Loss ◽  
Speech Recognition ◽  
Speech Signal ◽  
Mixed Model ◽  
Normal Hearing ◽  
Nonnative English Speakers ◽  
English Speakers ◽  
Masking Release ◽  
Accented English ◽  
Competing Speech

Background: Masking release for an English sentence-recognition task in the presence of foreign-accented English speech compared with native-accented English speech was reported in Calandruccio et al (2010a). The masking release appeared to increase as the masker intelligibility decreased. However, it could not be ruled out that spectral differences between the speech maskers were influencing the significant differences observed. Purpose: The purpose of the current experiment was to minimize spectral differences between speech maskers to determine how various amounts of linguistic information within competing speech affect masking release. Research Design: A mixed-model design with within-subject (four two-talker speech maskers) and between-subject (listener group) factors was conducted. Speech maskers included native-accented English speech and high-intelligibility, moderate-intelligibility, and low-intelligibility Mandarin-accented English. Normalizing the long-term average speech spectra of the maskers to each other minimized spectral differences between the masker conditions. Study Sample: Three listener groups were tested, including monolingual English speakers with normal hearing, nonnative English speakers with normal hearing, and monolingual English speakers with hearing loss. The nonnative English speakers were from various native language backgrounds, not including Mandarin (or any other Chinese dialect). Listeners with hearing loss had symmetric mild sloping to moderate sensorineural hearing loss. Data Collection and Analysis: Listeners were asked to repeat back sentences that were presented in the presence of four different two-talker speech maskers. Responses were scored based on the key words within the sentences (100 key words per masker condition). A mixed-model regression analysis was used to analyze the difference in performance scores between the masker conditions and listener groups. Results: Monolingual English speakers with normal hearing benefited when the competing speech signal was foreign accented compared with native accented, allowing for improved speech recognition. Various levels of intelligibility across the foreign-accented speech maskers did not influence results. Neither the nonnative English-speaking listeners with normal hearing nor the monolingual English speakers with hearing loss benefited from masking release when the masker was changed from native-accented to foreign-accented English. Conclusions: Slight modifications between the target and the masker speech allowed monolingual English speakers with normal hearing to improve their recognition of native-accented English, even when the competing speech was highly intelligible. Further research is needed to determine which modifications within the competing speech signal caused the Mandarin-accented English to be less effective with respect to masking. Determining the influences within the competing speech that make it less effective as a masker or determining why monolingual normal-hearing listeners can take advantage of these differences could help improve speech recognition for those with hearing loss in the future.

Measuring Mandarin Speech Recognition Thresholds Using the Method of Adaptive Tracking

2019 ◽  
Vol 62 (6) ◽  
pp. 2009-2017
Author(s):  
Yuxia Wang ◽  
Zhaoyu Lu ◽  
Xiaohu Yang ◽  
Chang Liu
Keyword(s):  
Speech Recognition ◽  
Adaptive Tracking ◽  
Mandarin Speech Recognition

Selecting the Optimal FM System for Children With Cochlear Implants

2008 ◽  
Vol 18 (1) ◽  
pp. 19-24
Author(s):  
Erin C. Schafer
Keyword(s):  
Speech Recognition ◽  
Cochlear Implants ◽  
Empirical Research ◽  
Background Noise ◽  
Signal To Noise Ratio ◽  
Evidence Based ◽  
Signal To Noise ◽  
Speech Processor ◽  
System Input ◽  
Optimal Type

Children who use cochlear implants experience significant difficulty hearing speech in the presence of background noise, such as in the classroom. To address these difficulties, audiologists often recommend frequency-modulated (FM) systems for children with cochlear implants. The purpose of this article is to examine current empirical research in the area of FM systems and cochlear implants. Discussion topics will include selecting the optimal type of FM receiver, benefits of binaural FM-system input, importance of DAI receiver-gain settings, and effects of speech-processor programming on speech recognition. FM systems significantly improve the signal-to-noise ratio at the child's ear through the use of three types of FM receivers: mounted speakers, desktop speakers, or direct-audio input (DAI). This discussion will aid audiologists in making evidence-based recommendations for children using cochlear implants and FM systems.

Sign in / Sign up

Export Citation Format

Share Document