用阿拉伯语大模型提升沙特语音识别准确率
Arabic ASR on the SADA Large-Scale Arabic Speech Corpus with Transformer-Based Models
- 基于Transformer的MMS 1B模型在苏丹数据集上微调
- 最佳结果：词错误率40.9%，字符错误率17.6%
- 适合做阿拉伯语语音识别或方言研究者参考
我们评估了多个先进的自动语音识别（ASR）模型在大规模阿拉伯语语音数据集SADA（沙特阿拉伯电视节目音频）上的表现，该数据集包含668小时高质量音频，涵盖多种方言和环境，尤其包含一个嘈杂子集，使识别更具挑战性。我们在SADA测试集上评估了模型性能，并研究了微调、语言模型以及噪声与去噪对性能的影响。结果表明，使用四元语法语言模型在SADA上微调后的MMS 1B模型，在测试集纯净部分取得了40.9%的词错误率（WER）和17.6%的字符错误率（CER），为当前最优表现。
原文摘要 · Abstract (English)
We explore the performance of several state-of-the-art automatic speech recognition (ASR) models on a large-scale Arabic speech dataset, the SADA (Saudi Audio Dataset for Arabic), which contains 668 hours of high-quality audio from Saudi television shows. The dataset includes multiple dialects and environments, specifically a noisy subset that makes it particularly challenging for ASR. We evaluate the performance of the models on the SADA test set, and we explore the impact of fine-tuning, language models, as well as noise and denoising on their performance. We find that the best performing model is the MMS 1B model finetuned on SADA with a 4-gram language model that achieves a WER of 40.9\% and a CER of 17.6\% on the SADA test clean set.
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。