开源语音识别评测平台，实现多语言长文本公平对比
Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Recognition Evaluation
- 构建可复现的开源评测平台，统一评估标准
- Conformer+Transformer组合最佳准确率，CTC/TDT更高效
- 适合研究者和工程师做模型选型与性能验证
我们提出 Open ASR Leaderboard，一个由学术界和工业界共同贡献的可复现基准平台。该平台在12个数据集上对比了86个开源与专有系统，涵盖英语短/长文本及多语言短文本任务。通过标准化词错误率（WER）和逆实时因子（RTFx），实现不同模型架构与工具包（如ESPNet、NeMo、SpeechBrain、Transformers）间的准确率-效率一致比较。实验发现，基于Conformer的编码器搭配Transformer解码器取得最优平均WER；而连接时序分类（CTC）与词元-时长转换器（TDT）解码器在RTFx上表现更优，更适合长文本与批量处理。所有代码与数据加载器均已开源，支持透明、可扩展的评估。本文方法旨在推动语音识别及其他任务的社区驱动评测。
原文摘要 · Abstract (English)
We present the Open ASR Leaderboard, a reproducible benchmarking platform with community contributions from academia and industry. It compares 86 open-source and proprietary systems across 12 datasets, with English short- and long-form and multilingual short-form tracks. We standardize word error rate (WER) and inverse real-time factor (RTFx) evaluation for consistent accuracy-efficiency comparisons across model architectures and toolkits (e.g., ESPNet, NeMo, SpeechBrain, Transformers). We observe that Conformer-based encoders paired with transformer-based decoders achieve the best average WER, while connectionist temporal classification (CTC) and token-and-duration transducer (TDT) decoders offer superior RTFx, making them better suited for long-form and batched processing. All code and dataset loaders are open-sourced to support transparent, extensible evaluation. We present our evaluation methodology to facilitate community-driven benchmarking in ASR and other tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。