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

筛选
英文 中文
Voistutor 2.0: A Speech Corpus with Phonetic Transcription for Pronunciation Evaluation of Indian L2 English Learners Voistutor 2.0:一个带音标的语音语料库,用于印度第二语言学习者的发音评价
Priyanshi Pal, Chiranjeevi Yarra, P. Ghosh
{"title":"Voistutor 2.0: A Speech Corpus with Phonetic Transcription for Pronunciation Evaluation of Indian L2 English Learners","authors":"Priyanshi Pal, Chiranjeevi Yarra, P. Ghosh","doi":"10.1109/O-COCOSDA202257103.2022.9997873","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997873","url":null,"abstract":"In computer assisted pronunciation training (CAPT), robust automatic models are critical for pronunciation assessment and mispronunciation detection and diagnosis (MDD). In the modelling, besides the audio data of second language (L2) learners, CAPT requires manually annotated ratings of overall pronunciation quality, and the MDD uses manually annotated phonetic transcriptions. Though the pronunciation quality and the mispronunciation are interdependent, to the best of our knowledge, none of the existing corpora contains both ratings and phonetic transcriptions. This could be due to the cost involved in obtaining phonetic transcriptions. However, a corpus with both kinds of information could benefit the researchers to obtain robust models by exploring the interdependencies. For addressing this, we develop voisTUTOR 2.0 corpus considering the existing voisTUTOR corpus referred to as voisTUTOR 1.0. We obtain phonetic transcriptions manually from a linguist for the entire Indian L2 learners’ English audio data (26529 utterances approximately 14 hours) in voisTUTOR 1.0 for which overall quality ratings and binary scores of factors influencing the pronunciation quality are available. A preliminary analysis of voisTUTOR 2.0 suggests that the phonetic errors correlated with the ratings and the binary scores indicating mispronunciations and phoneme quality.","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":"87 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":"124414816","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}
引用次数: 1
The Effect of Acoustic Features on Chinese EFL Learners’ Perception of English Accentual Prominence 声学特征对中国英语学习者英语重音突出感知的影响
Weizhong Zhang, Jian Gong, Xiaoli Ji, Yuhong Sun, Kai Sheng
{"title":"The Effect of Acoustic Features on Chinese EFL Learners’ Perception of English Accentual Prominence","authors":"Weizhong Zhang, Jian Gong, Xiaoli Ji, Yuhong Sun, Kai Sheng","doi":"10.1109/O-COCOSDA202257103.2022.9997972","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997972","url":null,"abstract":"A group of Chinese EFL learners participated in a perception experiment to identify English narrow focus words located in sentence-initial, -medial and -final positions. The results showed that learners had relatively high recognition accuracy for sentence-initial focus but performed significantly worse on sentence-final focus. Acoustic analysis revealed that, although native English speaker modulated a series of acoustic features significantly to realize accentual prominence for the focused words, only pitch range, peak F0, duration and intensity were used by Chinese EFL learners in the perception task, while F0 movement was largely ignored by them. These results indicated a strong native language influence for the Chinese EFL learners, that they transferred the Chinese focus realization habit in their English focus perception.","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":"127 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":"131414287","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}
引用次数: 0
Development of a High Quality Text to Speech System for LAO 高质量老挝语文本转语音系统的开发
Nguyen Thi Ngoc Anh, Nguyen Tien Thanh, Le Nhat Minh
{"title":"Development of a High Quality Text to Speech System for LAO","authors":"Nguyen Thi Ngoc Anh, Nguyen Tien Thanh, Le Nhat Minh","doi":"10.1109/O-COCOSDA202257103.2022.9997914","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997914","url":null,"abstract":"Text-to-speech, or speech synthesis, transforms text input to audio signal. It has a variety of applications and is a popular research area in the field of natural language and speech processing. However, many languages in the world have not been carefully researched. This paper presents the first study about high-quality Lao language speech synthesis; despite not receiving much attention, the language has a lot of future potential with applications in areas such as customer care, reading news, GPS navigation, etc. The database has been designed, processed, and used to build an end-to-end Lao text-to-speech model. Many tests are conducted and evaluated by native speakers, then chosen as the best model to demonstrate the potential of the proposed scheme to produce a high-quality Lao language synthesis.","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":"69 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":"122131021","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}
引用次数: 1
Philippine 菲律宾
J. Flavier, F. Ramos
{"title":"Philippine","authors":"J. Flavier, F. Ramos","doi":"10.1109/o-cocosda202257103.2022.9997989","DOIUrl":"https://doi.org/10.1109/o-cocosda202257103.2022.9997989","url":null,"abstract":": Nrf2 Function as Chemopreventive Compounds and Chemoresistance Nuclear factor erythroid 2 - related factor 2 (NRF2) is a transcription factor and the master regulator of antioxidant response. This mechanism protects against oxidative stress and promotes cell survival. The role of NRF2 in cancer prevention and treatment is complex due to its duality. NRF2 is required in protecting the healthy cells of the body by binding to the antioxidant response element (ARE) of its target genes and producing beneficial proteins, which enables cellular survival. However in some cancer cells, due to mutations or other cellular events NRF2 stays constitutively active, resulting in the promotion of cancer. A NRF2 inhibitor is needed to overcome the constitutive activation of NRF2 in cancer cells and for further research. This study focuses on developing compounds that could inhibit NRF2 function, and decrease cancer cell survival. Several NRF2 inhibitor compounds were developed and tested on p53 deficient cancer cell line H1299. H1299 cells were treated for 4 or 16 hours and the expression levels of NRF2 and its target genes were determined with Western blots. Surprisingly there was an increase in NRF2 protein levels, however most downstream proteins (NQO1, GCLM, P62) were downregulated, resulting in decreased cell survival. The current results suggest that not only NRF2 inhibition with current compounds prevent NRF2 degradation but also NRF2 function is decreased. Further research in needed to identify and characterize the inhibition of NRF2 function in cancer cells and also prevent cancer cells survival. ABSTRACT: Gantt charts are commonly used for visualizing execution traces in parallel processes. Examining and understanding these execution traces become important for performance debugging and general comprehension of program fluidity. As complexity grows we are faced with algorithms that create hundreds of processes raising the question of how to effectively display this data. The study addresses the question of whether the current methods for visualizing Gantt charts, specifically in relation to execution traces, are effective by conducting a user study using multiple choice questions and grid drawing questions. We conducted a pilot study in which subjects answered a series of question regarding visualization patterns in Gantt charts to investigate whether these visualization patterns were intuitive to users in their present form. We are currently in the process of analyzing the data from our pilot survey to be able to better narrow down the types of questions to use moving forward into the final study. We plan to run this study again at Supercomputing Conference 2018 to get participants with diverse programming backgrounds to effectively analyze if users can easily understand and interpret visualization patterns. ABSTRACT: The Philippines possesses a vibrant history of elitist power, martial law, and constant reach for improved governance. The influence o","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":"114953124","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}
引用次数: 8
Analysis of Layer-Wise Training in Direct Speech to Speech Translation Using BI-LSTM 基于BI-LSTM的直接语音到语音翻译的分层训练分析
Lalaram Arya, Ayush Agarwal, Jagabandhu Mishra, S. Prasanna
{"title":"Analysis of Layer-Wise Training in Direct Speech to Speech Translation Using BI-LSTM","authors":"Lalaram Arya, Ayush Agarwal, Jagabandhu Mishra, S. Prasanna","doi":"10.1109/O-COCOSDA202257103.2022.9997945","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997945","url":null,"abstract":"Speech-to-speech translation (S2ST) is the process of translation of speech from one language to another. Traditional S2ST systems follow a cascaded approach, where three modules automatics speech recognition (ASR), machine translation (MT), and text-to-speech translation (TTS) are concatenated to obtain the final translated speech utterance. The cascaded nature of the system results in the propagation of errors from one module to another. This, in turn, leads to the degradation in the overall performance of the S2ST task. With the evolution of the deep learning approaches to speech processing, many attempts have been made to perform end-to-end and direct speech-to-speech translation (DS2ST). But most of these approaches rely on language transcripts in one way or the other. In this work, we aim to perform the DS2ST task without using language transcripts. In this direction we have performed three experiments: First, we have investigated the direct learning of mapping function from source to target language with the increase in the number of utterances. Second, we have analyzed how the learning function improves with an increase in the number of Bi-LSTM layers. Third, we have observed how the system behaves with the unknown speakers (not used during training) during inference. From the experiments, it has been observed that with the increase in the number of utterances and layers, the quality of translation improves. And also, with a speaker and text-dependent training of approximately 4.4 hrs of speech, the model can generate the target language utterance even for unknown speakers. Though the generated utterance quality is not that good, but intelligent to some extent to be perceived.","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":"63 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":"115029761","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}
引用次数: 0
Text to Speech System for Lambani - A Zero Resource, Tribal Language of India 兰巴尼语的文本到语音系统-零资源,印度部落语言
Ashwini Dasare, K. Deepak, S. Prasanna, K. Samudravijaya
{"title":"Text to Speech System for Lambani - A Zero Resource, Tribal Language of India","authors":"Ashwini Dasare, K. Deepak, S. Prasanna, K. Samudravijaya","doi":"10.1109/O-COCOSDA202257103.2022.9997838","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997838","url":null,"abstract":"A Text to Speech (TTS) system empowers illiterate people by speaking, in their native language, the information available in electronic media. Speech data and corresponding transcript is essential for the development of a TTS system. Most tribal languages in India neither have script nor written literature. In this paper, we present our effort to build a TTS system for Lambani, a tribal language spoken by a group of nomadic people living in several regions of India. We generated a text corpus of about 3000 Lambani sentences, written in the script of Kannada language. These sentences were read by a literate Lambani speaker in a recording studio. The Lambani TTS system was developed by adapting the Nvidia’s Tacotron2 model, pretrained with English speech, following the transfer learning approach. We implemented two versions of the Lambani TTS system. The first version uses HiFi-GAN vocoder and the second uses the WaveGlow vocoder. We evaluated the two versions of the TTS system using both objective and subjective measures. Both Perceptual Evaluation of Speech Quality score and Mean Opinion Score were higher for the Lambani TTS system that used WaveGlow vocoder.","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":"35 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":"121166021","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}
引用次数: 0
Spanish Stops And Their Allophones Produced By Proficient Mandarin Learners Of Spanish 熟练的西班牙语普通话学习者产生的西班牙语停顿及其音素
Xiaotong Xi, Peng Li
{"title":"Spanish Stops And Their Allophones Produced By Proficient Mandarin Learners Of Spanish","authors":"Xiaotong Xi, Peng Li","doi":"10.1109/O-COCOSDA202257103.2022.9997932","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997932","url":null,"abstract":"This study examined the pronunciation of Spanish stops by Mandarin learners of Spanish, with the focus on the realization of voiced stops, voiceless stops, and allophonic spirants of voiced stops in isolated words and in running speech. We recruited 10 Mandarin-speaking learners with high-proficiency Spanish skills and 6 native Peninsular Spanish speakers. They read words containing the target sounds and a short paragraph. Voice onset time was measured for the stops and the manner of articulation was coded for the allophones of voiced stops. The results showed that Mandarin learners of Spanish could distinguish the two types of stops. However, their production of voiceless stops in isolated words and the voiced stops in the paragraph were significantly different from native speakers. Moreover, most Mandarin speakers could not produce spirantized allophone of voiced stops as much as native speakers did in both reading tasks. These results indicate that the pronunciation of phonemes in a second language and the acquisition of nonnative allophonic phonemes are still challenging even for high-proficiency learners.","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":"26 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":"115836491","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}
引用次数: 0
Program with Session Chair 主持人主持的节目
{"title":"Program with Session Chair","authors":"","doi":"10.1109/o-cocosda202257103.2022.9997951","DOIUrl":"https://doi.org/10.1109/o-cocosda202257103.2022.9997951","url":null,"abstract":"","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":"62 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":"133991624","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}
引用次数: 0
Harvard-NGSL Sentences for English Learner Speech Corpora 哈佛-英语学习者口语语料库中的ngsl句子
Kakeru Yazawa
{"title":"Harvard-NGSL Sentences for English Learner Speech Corpora","authors":"Kakeru Yazawa","doi":"10.1109/O-COCOSDA202257103.2022.9998002","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9998002","url":null,"abstract":"This paper introduces a set of phonemically balanced materials for English learner speech corpora named the Harvard-NGSL sentences. They are a subset of the IEEE Harvard sentences, a collection of 720 phonemically balanced short sample phrases. Of these, 50 sentences that comprise core high-frequency vocabulary words for second language English learners were selected based on the New General Service List (NGSL). The resultant Harvard-NGSL sentences are thus learner-oriented, unlike most other phonemically balanced materials which contain low-frequency and low-familiarity words that are of limited use for learners. An analysis based on the CMU Pronouncing Dictionary found that the sentences show a similar frequency distribution of all phonemes to that of the original Harvard sentences except/3/, which is extremely uncommon in English and thus of minimal importance to learners. The advantages of the proposed materials and directions for future research are discussed.","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":"307 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":"129335589","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}
引用次数: 0
Towards The Development Of Accent Conversion Model For (L1)Bengali Speaker Using Cycle Consistent Adversarial Network (Cyclegan) 基于循环一致对抗网络(Cyclegan)的(L1)孟加拉语说话人口音转换模型研究
S. Chandra, Puja Bharati, S. Mandal
{"title":"Towards The Development Of Accent Conversion Model For (L1)Bengali Speaker Using Cycle Consistent Adversarial Network (Cyclegan)","authors":"S. Chandra, Puja Bharati, S. Mandal","doi":"10.1109/O-COCOSDA202257103.2022.9997837","DOIUrl":"https://doi.org/10.1109/O-COCOSDA202257103.2022.9997837","url":null,"abstract":"The goal of foreign accent conversion (FAC) is to create a new voice with the voice identity of a given second-language (L2) speaker but with a native (L1) accent. The main motivation for this work is to solve the problem of Indians understanding foreigners’ accents and foreigners understanding Indian accents. We propose a non-parallel accent conversion (AC) method that can learn mapping from source to target without the use of parallel speech data. The proposed method is particularly advantageous in that it does not require any additional data or an alignment procedure. A cycle-consistent adversarial network is used in our proposed method. This enables the identification of an optimal pseudo pair. from nonparallel dataset. We applied our method on a speech data contained only American English and English speech recorded by Bengali speaker. As per our knowledge this is the first attempt to convert the accent native (L1 American English speaker) and nonnative (L1 Bengali speaker). We trained our model under disadvantageous condition (non-parallel data). Result shows that this approach will be useful in accent conversion field.","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":"177 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":"129253569","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}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书