Enabling ASR for Low-Resource Languages: A Comprehensive Dataset Creation Approach
American University of Armenia(亚美尼亚美国大学);Nvidia(英伟达)
Ara Yeroyan, Nikolay Karpov
备注:
13 pages, 10 figures (including ablation studies), to be published in 2024 IEEE Spoken Language Technology Workshop. Additionally, the associated software package can be accessed at (https://pypi.org/project/vac-aligner/) for practical applications and further development
In recent years, automatic speech recognition (ASR) systems have significantly improved, especially in languages with a vast amount of transcribed speech data. However, ASR systems tend to perform poorly for low-resource languages with fewer resources, such as minority and regional languages. This study introduces a novel pipeline designed to generate ASR training datasets from audiobooks, which typically feature a single transcript associated with hours-long audios. The common structure of these audiobooks poses a unique challenge due to the extensive length of audio segments, whereas optimal ASR training requires segments ranging from 4 to 15 seconds. To address this, we propose a method for effectively aligning audio with its corresponding text and segmenting it into lengths suitable for ASR training. Our approach simplifies data preparation for ASR systems in low-resource languages and demonstrates its application through a case study involving the Armenian language. Our method, which is "portable" to many low-resource languages, not only mitigates the issue of data scarcity but also enhances the performance of ASR models for underrepresented languages.
LRSpeech: Extremely Low-Resource Speech Synthesis and Recognition
Institute for Interdisciplinary Information Sciences, Tsinghua University(清华大学交叉信息研究院);Zhejiang University(浙江大学);Microsoft Research Asia(微软亚洲研究院);Microsoft Azure Speech(微软 Azure 语音)
语音合成(text to speech,TTS)和识别(自动语音识别,ASR)是重要的语音任务,并且需要大量的文本和语音对用于模型训练。然而,世界上有超过6,000种语言,大多数语言都缺乏语音训练数据,这在为资源极低的语言构建TTS和ASR系统时带来了巨大的挑战。在本文中,我们开发了LRSpeech,一个在极低资源环境下的TTS和ASR系统,它可以支持稀有语言,低数据成本。LRSeep包括三个关键技术:1)对资源丰富的语言进行预训练,并对资源不足的语言进行微调; 2)TTS和ASR之间的双重转换,以迭代方式提高彼此的准确性; 3)知识蒸馏,以定制高质量目标说话人语音的TTS模型,并改进多个语音的ASR模型。我们进行实验的实验语言(英语)和一个真正的低资源的语言(立陶宛语),以验证LRSeech的有效性。实验结果表明,LRSeep 1)在合成语音的可懂度(超过98%的可懂度)和自然度(超过3.5的平均意见得分(MOS))方面实现了高质量的TTS,满足工业部署的要求,2)实现了有希望的ASR识别准确率,3)最后但并非最不重要的是,使用极低资源的训练数据。我们还利用不同数量的数据资源对LRSeech进行全面分析,为行业部署提供有价值的见解和指导。我们目前正在将LRSpeech部署到商业化的云语音服务中,以支持更多稀有语言的TTS。
SSVD: Structured SVD for Parameter-Efficient Fine-Tuning and Benchmarking under Domain Shift in ASR
Department of Electrical Engineering KU Leuven Leuven, Belgium(电气工程系 卡尔·冯·特里尔大学 布鲁塞尔,比利时);Language Technologies Institute Carnegie Mellon University Pittsburgh, PA, USA(语言技术研究所 哈佛大学 贝尔蒙特,美国)
MNV-17: A High-Quality Performative Mandarin Dataset for Nonverbal Vocalization Recognition in Speech
South China University of Technology(华南理工大学);The Hong Kong Polytechnic University(香港理工大学);Tongji University(同济大学);Shanghai Jiao Tong University(上海交通大学);The Chinese University of Hong Kong(香港中文大学);Foshan University(佛山大学)
Automatic Creation of Text Corpora for Low-Resource Languages from the Internet: The Case of Swiss German
iCoSys Institute, University of Applied Sciences and Arts Western Switzerland(瑞士西部应用科学与艺术大学iCoSys研究所);University of Fribourg(弗里堡大学);Swisscom AG(瑞士电信公司)
WenetSpeech-Yue: A Large-scale Cantonese Speech Corpus with Multi-dimensional Annotation
Institute of Artificial Intelligence (TeleAI), China Telecom(人工智能研究院(TeleAI),中国电信);Beijing AISHELL Technology Co., Ltd.(北京AISHELL科技有限公司);WeNet Open Source Community(WeNet开源社区);Hong Kong University of Science and Technology(香港理工大学)
A Comprehensive Study on the Effectiveness of ASR Representations for Noise-Robust Speech Emotion Recognition
Graduate School of Informatics, Nagoya University(国立 Nagoya 大学信息科学研究生院);Faculty of Data Science, City University of Macau(澳门城市大学数据科学学院);Information Technology Center, Nagoya University(国立 Nagoya 大学信息技术中心)
Avaya Inc.(亚美亚公司);Wrocław University of Science and Technology(弗罗茨瓦夫科技大学);Johns Hopkins University(约翰斯·霍普金斯大学);Poznań University of Technology(波兹南理工大学)
Lahore University of Management Sciences(拉瓦尔大学管理科学学院);University of Michigan - Ann Arbor(密歇根大学安娜堡分校);EMBL European Bioinformatics Institute(欧洲生物信息学研究所)
Investigation of End-To-End Speaker-Attributed ASR for Continuous Multi-Talker Recordings
Microsoft(微软)
近年来,针对单声道重叠语音,提出了一种端到端(E2 E)说话人属性自动语音识别(SA-ASR)模型,作为说话人计数、语音识别和说话人识别的联合模型。它显示了有前途的结果,由不同数量的扬声器的模拟语音混合物。然而,该模型需要说话人轮廓的先验知识来进行说话人识别,这极大地限制了该模型的应用。在本文中,我们扩展了以前的工作,解决的情况下,没有扬声器配置文件。具体来说,我们使用E2 E SA-ASR模型的内部扬声器表示来进行扬声器计数和聚类,以将扬声器库存中缺少配置文件的扬声器的话语进行diarize。我们还对E2 E SA-ASR训练的参考标签提出了一个简单的修改,这有助于很好地处理连续的多说话者录音。我们对原始E2 E SA-ASR和单耳LibriCSS数据集上的所提出的方法进行了全面的调查。与原始的E2 E SA-ASR与相关的扬声器配置文件相比,所提出的方法实现了接近的性能,没有任何先验扬声器知识。我们还表明,在E2 E SA-ASR模型中的源-目标注意力提供了关于假设的开始和结束时间的信息。
MF-AED-AEC: Speech Emotion Recognition by Leveraging Multimodal Fusion, Asr Error Detection, and Asr Error Correction
Graduate School of Informatics, Nagoya University(名古屋大学信息学研究科);School of Computer Science and Technology, Hainan University(海南大学计算机科学与技术学院);Information Technology Center, Nagoya University(名古屋大学信息技术中心)
Institute of Health Informatics, University of Minnesota(明尼苏达大学健康信息学研究所);Biomedical Informatics and Medical Education, University of Washington(华盛顿大学生物医学信息学与医学教育系);Department of Pharmaceutical Care & Health Systems, University of Minnesota(明尼苏达大学药物护理与卫生系统系)
Discovering Phonetic Inventories with Crosslingual Automatic Speech Recognition
Center of Language and Speech Processing, The Johns Hopkins University(约翰斯·霍普金斯大学语言与语音处理中心);Human Language Technology Center of Excellence, The Johns Hopkins University(约翰斯·霍普金斯大学人类语言技术卓越中心);Multimedia Computing Group, Delft University of Technology(代尔夫特理工大学多媒体计算组);Department of Electrical and Computer Engineering, University of Illinois(伊利诺伊大学电气与计算机工程系)
TG-ASR: Translation-Guided Learning with Parallel Gated Cross Attention for Low-Resource Automatic Speech Recognition
Dept. Computer Science and Information Engineering, National Taiwan Normal University, Taiwan(计算机科学与信息工程系,台湾国立台湾师范大学);Institute of Computer Science, Academia Sinica, Taiwan(学术院计算机科学研究所,台湾);United Link Co., Ltd., Taiwan(台湾联合链接有限公司)