2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)最新文献
{"title":"Analysis on the Interference Of Chinese /N/-/L/ Confusion to Japanese /N/-/R/ Discrimination by Chinese JFL Learners","authors":"Cenyu Xiang, Tianxiang Cao, Yanlong Zhang","doi":"10.1109/O-COCOSDA202257103.2022.9997970","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997970","url":null,"abstract":"Negative transfer occurs in the discrimination of similar phonemes between L1 and L2. Few works have been done on the confusion of Japanese /n/-/r/ by the Chinese JFL learners whose native dialect has the allophone of /n/-/l/, which is similar to Japanese /r/. Aiming at better understanding of the interference of L1 Mandarin /n/-/l/ on L2 Japanese /n/-/r/, we analyzed the effect of phonetic context and speech rate on the perceptual discrimination by 23 Chinese JFL learners of beginners. Participants identified Chinese /na/-/la/, /ni/-/li/ and Japanese /na/-/ra/ and /ni/-/ri/ in monosyllable words, fast sentences and slow sentences. Results showed that phonetic contexts influenced the discrimination of both L1 /n/-/l/ and L2 /n/-/r/. However, speech rate had little effect on the L1 contrast, while significantly affected the L2 contrast. The results bring more specific insights into the understanding of the interference of L1 allophone to the similar L2 phoneme discrimination.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122783772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Binh Van Duong, An Trong Nguyen, Chien Nhu Ha, Hong-Hanh Thi Duong, Myky Tran, Trong-Hop Do
{"title":"UIT-VLFC: Vietnamese Lipstick Feedbacks Corpus","authors":"Binh Van Duong, An Trong Nguyen, Chien Nhu Ha, Hong-Hanh Thi Duong, Myky Tran, Trong-Hop Do","doi":"10.1109/O-COCOSDA202257103.2022.9997964","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997964","url":null,"abstract":"Sentiment analysis for Vietnamese text can be applied in business fields such as sales, entertainment, advertisement, etc. Especially in the E-commerce era, thousands of products are sold online daily, followed by millions of users’ comments on the E-commerce platforms. For that reason, this research introduces the UIT-VLFC (UIT-Vietnamese Lipstick Feedbacks Corpus), a data set of users’ comments of lipstick products on E-commerce platforms. This aims to contribute a data set to the Vietnamese sentiment analysis field, even the sentiment-analysis recommendation system problem. The research also gives some experimental results on machine learning methods (Multinomial Naïve Bayes, Random forests, Logistic regression, Support vector machine (SVM)) and deep learning method (Bidirectional-Long short-term memory), etc. in order to assess the UIT-VLFC. The experimental results show that the UIT-VLFC has lots of implicit potentials that can be explored for many applications.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"343 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115889690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tzu-Han Lin, Kuan-lin Lee, Hsin-Yun Chung, Fu-Hai Frank Wu, Jui-Chu Li, Tung-Lung Li, Shih-Lung Lo, Yi-Wen Liu, Jason J. S. Chang
{"title":"Toward Automatic Generation of Transcript From Spoken Lectures: The “Dream of the Red Chamber” Series","authors":"Tzu-Han Lin, Kuan-lin Lee, Hsin-Yun Chung, Fu-Hai Frank Wu, Jui-Chu Li, Tung-Lung Li, Shih-Lung Lo, Yi-Wen Liu, Jason J. S. Chang","doi":"10.1109/O-COCOSDA202257103.2022.9997890","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997890","url":null,"abstract":"We present a corpus of Mandarin literature lectures. The recording device, environment, and the topics covered in the lectures are described briefly; then, we developed speech recognition and text simplification approaches for automatic transcription of the lectures. Given that the size of this corpus is relatively small, we applied transfer learning on AISHELL-1 by varying the number of transferred layers and fine-tuning the learning rates to improve the accuracy of speech recognition. Experimental results showed that a character error rate (CER) of 15.83% could be obtained. Additionally, by reducing the perplexity of the language model and the number of out-of-vocabulary words, the CER improved by another 0.29%. Further, to improve the fluency of transcription, we chose to deal with punctuation and pleonasm. Accuracy of 81% and 91.5% were reported by professional judges on adding punctuation and deleting pleonasm, respectively.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132284373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Neural Network Models For User Attribute Extraction From Dialogues","authors":"Son Dang, Pham Quang Nhat Minh","doi":"10.1109/O-COCOSDA202257103.2022.9997870","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997870","url":null,"abstract":"In this paper, we present an approach to extracting user attributes represented in the form of Subject-Predicate-Object (SPO) triplets from utterances in dialogues. We treated user attribute extraction as a two-tasks process: 1) Predicate Classification and 2) Entity Generation. The predicate classification task is to determine predicates triggered in a sentence, and the entity generation task will generate related subjects and objects for each triggered predicate. Since multiple predicates may be triggered in one sentence, we formalized the predicate classification task as a multi-label classification problem. We fine-tuned a pre-trained BERT model to solve the problem. We formalized the entity generation task as a sequence-to-sequence problem and fine-tuned the T5 model on the task-specific data to create an entity generation model. Experimental results showed that our proposed method out-performs a state-of-the-art baseline model by a large margin.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124164988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Effects Of The Syntactic Structure On The Productivity Of Tone Sandhi Rules: In The Case Of Xiamen Dialect","authors":"Yiying Hu, Hui Feng","doi":"10.1109/O-COCOSDA202257103.2022.9997918","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997918","url":null,"abstract":"The productivity of tone sandhi rules in Xiamen Dialect has been explored in recent studies, but the syntactic structure is rarely taken as a variate. This study investigates effects of the syntactic structure on the productivity of tone sandhi rules from a diachronic perspective. Eight native speakers of Xiamen Dialect were invited to read real words and wug words in different syntactic structures. Findings are as follows: (1) Syntactic structures exerted statistically significant effects on the productivity of tone sandhi rules in Xiamen Dialect. Participants articulated more correct sandhi tones for sandhi targets in wug words with clear syntactic structures. (2) With the diachronic development, the effects of the syntactic structure on the productivity of tone sandhi rules in Xiamen Dialect became more obvious as adolescents tended to articulate citation tones for head verbs.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127164673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Syntactic Complexity In Narrative Speech Produced By Prelingually Deaf Mandarin-Speaking Children With Cochlear Implants","authors":"Yu Jue, Sun Han, Dong Zhaoqi","doi":"10.1109/O-COCOSDA202257103.2022.9997882","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997882","url":null,"abstract":"This paper examined the syntactic production performance of Mandarin-speaking children with cochlear implants (CIs) in narrative speech. The main purpose was to investigate how CI children differed from their normal-hearing (NH) peers in syntactic proficiency. The results showed that compared with their NH peers: 1) CI children were generally less productive in narrative speech, as they failed to produce more and longer utterances to potentially deliver complex and rich information; 2) used more simple-structured utterances with less variant length and preferred to apply coordination than subordination structure, in terms of T-unit measures and utterance types. 3) produced cognitively less complex utterances in terms of the dependency distance. Finally, earlier implantation had a positive impact on the syntactic production performance of the CI children while longer device use may not have such significant contribution.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121257890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Soeun Park, Jihye Chun, Mi Hyun Kim, Hyunjoo Lee, S. Lee, Sunhee Kim
{"title":"Designing A Korean French-Learners’ Speech Corpus (KFLSC) for Spoken Language Assessment","authors":"Soeun Park, Jihye Chun, Mi Hyun Kim, Hyunjoo Lee, S. Lee, Sunhee Kim","doi":"10.1109/O-COCOSDA202257103.2022.9997999","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997999","url":null,"abstract":"This paper presents how to design a French speech corpus produced by Korean learners (Korean French-learners’ Speech Corpus: KFLSC), which aims to provide an AI training database for spoken language assessment within the Korean AI Hub Project. The corpus is required to include audio recording, their corresponding transcription, metadata, assessment labeling, and error annotations. Both reading speech and spontaneous speech will be recorded by using our scripts and instruction prompts. The transcription of audio files will be performed both automatically and manually. Detailed speech assessment and scoring information are provided along with proposed pronunciation and spoken language assessment rubrics. We also suggest tagging guidance on pronunciation and spoken language errors. This study will contribute to building a large-scale French learner corpus for French spoken language assessment and the education of French as a foreign language, and to the development of automatic speech assessment and computer-aided language learning technologies.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133558019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Nguyen, T. Nguyen, Quoc Truong Do, Tuan L Nguyen
{"title":"End-to-end named entity recognition for Vietnamese speech","authors":"T. Nguyen, T. Nguyen, Quoc Truong Do, Tuan L Nguyen","doi":"10.1109/O-COCOSDA202257103.2022.9997862","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997862","url":null,"abstract":"One of the first steps in comprehending natural or spoken language is named-entity recognition. It plays a critical role in natural language processing applications such as text clustering, subject detection, topic detection, and text summarization. The most common techniques in information extraction using named-entity recognition can be applied to building automatic question-answering systems, semantic web technology, automatic translation machines, and so on. For input text with correct formatting, studies on the named entity recognition (NER) task have achieved excellent and nearly human-equal results. However, for the output documents of the speech recognition system (ASR), the normative signs of the text, such as punctuation and uppercase, no longer exist, causing difficulties for the researchers. In this paper, we present the process of building a Vietnamese speech dataset for the NER task and propose a new end-to-end approach for the NER of Vietnamese speech. In addition, we combine the multi-task learning model with the punctuation and uppercase (CaPu) recovery model and demonstrate the combination’s effectiveness when documenting the approximately 5% improvement in the F1 score.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134101975","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Creakiness Judgments by Burmese and Vietnamese Speakers","authors":"J. Villegas, Seunghun J. Lee","doi":"10.1109/O-COCOSDA202257103.2022.9997908","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997908","url":null,"abstract":"Vietnamese and Burmese listeners judged as creaky or not words in Du’an Zhuang, an unintelligible language for both groups. Although these languages differ in their tonal system, they use creakiness for phonemic contrast. The ratings of the two groups were similar, except in two non-creaky tones where the Vietnamese ratings were more likely to be non-creaky than the Burmese ratings. In terms of accuracy, the Vietnamese group was better doing these classifications. The differences between the two groups could indicate an influence of linguistic background in the determination of the creakiness of unintelligible words, however, it is also possible that the lack of control in the experimental setup (data was collected via an online survey) could have had a large effect on these results.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130314389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Peiwen Li, Enze Su, Jia Li, Siqi Cai, Longhan Xie, Haizhou Li
{"title":"Esaa: An Eeg-Speech Auditory Attention Detection Database","authors":"Peiwen Li, Enze Su, Jia Li, Siqi Cai, Longhan Xie, Haizhou Li","doi":"10.1109/O-COCOSDA202257103.2022.9997944","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997944","url":null,"abstract":"Humans are able to listen to a particular sound source in a noisy environment, an ability which is referred to as the cocktail party effect. Auditory attention detection (AAD) sheds light on the neural mechanisms of the cocktail party problem and enables neuro-steered hearing prostheses. In this paper, we build a database for AAD research, which consists of competing speech stimuli and associated human neural responses, i.e, electroencephalography (EEG) recordings, namely EEG-Speech AAD (ESAA) database. This is the first AAD database with speech stimuli in a tonal language (Mandarin), which contains 12.7 hours of data collected from 20 subjects. Moreover, we develop an AAD baseline as a reference model for decoding which speech stream a listening subject is attending to (speaker attention detection), and a baseline for decoding which spatial locus a listening subject is attending to (speaker locus attention detection) on the ESAA database. We achieve the accuracy of 84.6% and 84.3% for speaker and speaker locus attention detection with 64-channel and 1-second decision window, respectively.","PeriodicalId":220366,"journal":{"name":"2022 25th Conference of the Oriental COCOSDA International Committee for the Co-ordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132781467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}