本文提出 ML-SUPERB 2.0 基准,在多种下游模型、微调设置和高效适配方法下评估预训练 SSL 与监督语音模型,揭示模型设计、语言及数据集对多语言 ASR 性能的显著影响。
摘要:
ML-SUPERB 在语言识别和自动语音识别(ASR)任务上评估自监督学习(SSL)模型。该基准将模型视为特征提取器,并使用单一的浅层下游模型,该模型可针对下游任务进行微调。然而,真实世界的使用场景可能需要不同的配置。本文提出了 ML-SUPERB 2.0,这是一个新的基准,用于跨下游模型、微调设置和高效模型适配方法评估预训练 SSL 和监督语音模型。我们发现,相比 ML-SUPERB 的设置,性能有所提升。但是,性能取决于下游模型的设计。此外,我们发现不同语言和数据集之间存在巨大的性能差异,这表明需要更有针对性的方法来提升多语言 ASR 性能。
英文摘要:
ML-SUPERB evaluates self-supervised learning (SSL) models on the tasks of language identification and automatic speech recognition (ASR). This benchmark treats the models as feature extractors and uses a single shallow downstream model, which can be fine-tuned for a downstream task. However, real-world use cases may require different configurations. This paper presents ML-SUPERB~2.0, which is a new benchmark for evaluating pre-trained SSL and supervised speech models across downstream models, fine-tuning setups, and efficient model adaptation approaches. We find performance improvements over the setup of ML-SUPERB. However, performance depends on the downstream model design. Also, we find large performance differences between languages and datasets, suggesting the need for more targeted approaches to improve multilingual ASR performance.
NVV-SuperBench: Beyond Words, Beyond Quality-Benchmarking Nonverbal Vocalizations in Speech Generation
Nanjing University(南京大学);The Hong Kong University of Science and Technology(香港科学与技术大学);The Chinese University of Hong Kong(香港中文大学);University of Science and Technology Beijing(北京科技大学);Northwestern Polytechnical University(西北工业大学);Shanghai Jiao Tong University(上海交通大学);National Taiwan University(国立台湾大学)
One Law, Many Languages: Benchmarking Multilingual Legal Reasoning for Judicial Support
University of Bern(伯尔尼大学);University of Copenhagen(哥本哈根大学);Bern University of Applied Sciences(伯尔尼应用科学大学);University of Fribourg(弗里堡大学);Stanford University(斯坦福大学)
PRiSM: Benchmarking Phone Realization in Speech Models
CMU(卡内基梅隆大学);Gwangju Institute of Science and Technology(光州科学技术院);UT Austin(得克萨斯大学奥斯汀分校);LMU Munich(慕尼黑路德维希-马克西米利安大学);UC Berkeley(伯克利加州大学);NVIDIA(英伟达);UBC(不列颠哥伦比亚大学)
GEMv2: Multilingual NLG Benchmarking in a Single Line of Code
Allen Institute for AI(艾伦人工智能研究所);Amazon Alexa AI(亚马逊Alexa人工智能部门);Bangladesh University of Engineering and Technology(孟加拉国工程技术大学);Carnegie Mellon University(卡内基梅隆大学);Charles University(查理大学)
SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring
Shanghai Jiao Tong University(上海交通大学);Peking University(北京大学);The Hong Kong University of Science and Technology(香港科技大学);Douyin Group(抖音集团);University of Chinese Academy of Sciences(中国科学院大学);National University of Singapore(新加坡国立大学);Monash University(莫纳什大学)
Multi-EuP: The Multilingual European Parliament Dataset for Analysis of Bias in Information Retrieval
School of Computing & Information Systems, The University of Melbourne(墨尔本大学计算与信息系统学院);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
MULTITuDE: Large-Scale Multilingual Machine-Generated Text Detection Benchmark
Kempelen Institute of Intelligent Technologies(肯佩伦智能技术研究院);The Pennsylvania State University(宾夕法尼亚州立大学);University of Mississippi(密西西比大学);MIT Lincoln Laboratory(麻省理工学院林肯实验室)
KyrgyzLLM-Bench: Benchmarking Kyrgyz Language Understanding
Institute of IT, Kyrgyz State Technical University named after I. Razzakov(以I. 拉扎科夫命名的吉尔吉斯国立技术大学信息技术学院);School of Computer Science, University of Windsor(温莎大学计算机科学学院);St. Petersburg Department of the Steklov Math. Institute, RAS St. Petersburg State University(俄罗斯科学院斯捷克洛夫数学研究所圣彼得堡分部,圣彼得堡国立大学)
UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment
University of Bath(巴斯大学);Cardiff University(卡迪夫大学);National University Philippines(菲律宾国家大学);Bielefeld University(比勒菲尔德大学);University of Exeter(埃克塞特大学);University of Gothenburg(哥德堡大学);UCLouvain(列日大学);MBZUAI(马克斯·普朗克智能系统研究所);Brigham Young University(Brigham Young 大学);INESC-ID Lisboa(里斯本 INESC-ID);Instituto Universitário de Lisboa (ISCTE-IUL)(里斯本大学学院(ISCTE-IUL));Universitat Pompeu Fabra(庞培法华大学);National Research Council, Canada(加拿大国家研究委员会);King Abdulaziz University(国王阿卜杜勒-阿齐兹大学)
DimStance: Multilingual Datasets for Dimensional Stance Analysis
University of Göttingen(弗赖堡大学);Yuan Ze University(元智大学);Imperial College London(伦敦帝国理工学院);University of Pretoria(南非普里蒂亚大学);National Yang Ming Chiao Tung University;Maseno University(马斯诺大学);Central Police University Taiwan(台湾中央警察大学);Yunnan University(云南大学);Bayero University Kano(卡诺巴耶罗大学);National Research Council Canada(加拿大国家研究委员会)
AfriMTEB and AfriE5: Benchmarking and Adapting Text Embedding Models for African Languages
University of Toronto(多伦多大学);Mila - Quebec AI Institute, McGill University(魁北克AI研究院,麦吉尔大学);Saarland University, Saarland Informatic Campus(萨尔兰大学,萨尔兰信息学院);Canada CIFAR AI Chair(加拿大CIFAR人工智能主席)
Consolidating and Developing Benchmarking Datasets for the Nepali Natural Language Understanding Tasks
Information and Language Processing Research Lab (ILPRL)(信息与语言处理研究实验室);Kathmandu University(加德满都大学);Institute for Research and Innovation in Intelligent Systems (IRIIS)(智能系统研究与创新研究院)