超越正字法：阿拉伯语短元音与方言语音的自动恢复
Beyond Orthography: Automatic Recovery of Short Vowels and Dialectal Sounds in Arabic
AI总结：
提出方言语音与元音化恢复框架，并构建 ArabVoice15 测试集，在极少阿拉伯语方言训练数据下显著降低字符错误率。
摘要：
本文提出了一种新颖的 Dialectal Sound and Vowelization Recovery（方言语音与元音化恢复）框架，旨在识别语音多样性强、方言丰富的语言中的借音和方言语音，其范围超越了标准正字法语音集。所提出框架使用带或不带连续预训练自监督表示的量化输入序列。我们以数据有限的阿拉伯语验证了该流程的有效性；阿拉伯语是一种方言丰富的语言，包含超过22种主要方言。方言阿拉伯语在语音上正确的转写语音资源十分稀缺。因此，我们推出 ArabVoice15，这是首个此类精心构建的测试集，涵盖来自15个阿拉伯国家的5小时方言语音，并配有语音上准确的转写，包括借音和方言特有语音。我们详细描述了标注指南，并分析了方言混淆对。广泛评估同时包含主观的人类感知测试和客观指标。在三个测试集上报告的实证结果表明，仅使用一个半小时训练数据，我们的模型在 ArabVoice15 上相比基线将字符错误率降低约7%。
英文摘要：
This paper presents a novel Dialectal Sound and Vowelization Recovery framework, designed to recognize borrowed and dialectal sounds within phonologically diverse and dialect-rich languages, that extends beyond its standard orthographic sound sets. The proposed framework utilized a quantized sequence of input with(out) continuous pretrained self-supervised representation. We show the efficacy of the pipeline using limited data for Arabic, a dialect-rich language containing more than 22 major dialects. Phonetically correct transcribed speech resources for dialectal Arabic are scarce. Therefore, we introduce ArabVoice15, a first-of-its-kind, curated test set featuring 5 hours of dialectal speech across 15 Arab countries, with phonetically accurate transcriptions, including borrowed and dialect-specific sounds. We described in detail the annotation guideline along with the analysis of the dialectal confusion pairs. Our extensive evaluation includes both subjective -- human perception tests and objective measures. Our empirical results, reported with three test sets, show that with only one and half hours of training data, our model improve character error rate by ~ 7\% in ArabVoice15 compared to the baseline.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
