In this paper, we analyse the error patterns of the raw waveform acoustic models in TIMIT's phone recognition task. Our analysis goes beyond the conventional phone error rate (PER) metric. We categorise the phones into three groups: {affricate, diphthong, fricative, nasal, plosive, semi-vowel, vowel, silence}, {consonant, vowel+, silence}, and {voiced, unvoiced, silence} and, compute the PER for each broad phonetic class in each category. We also construct a confusion matrix for each category using the substitution errors and compare the confusion patterns with those of the Filterbank and Wav2vec 2.0 systems. Our raw waveform acoustic models consists of parametric (Sinc2Net) or non-parametric CNNs and Bidirectional LSTMs, achieving down to 13.7%/15.2% PERs on TIMIT Dev/Test sets, outperforming reported PERs for raw waveform models in the literature. We also investigate the impact of transfer learning from WSJ on the phonetic error patterns and confusion matrices. It reduces the PER to 11.8%/13.7% on the Dev/Test sets.

翻译：本文分析了原始波形声学模型在TIMIT音素识别任务中的错误模式。我们的分析超越了传统的音素错误率（PER）度量标准。我们将音素分为三组：{塞擦音、双元音、擦音、鼻音、塞音、半元音、元音、静音}，{辅音、元音+、静音}，以及{浊音、清音、静音}，并计算每个类别中每个广义语音类别的PER。我们还利用替换错误为每个类别构建了混淆矩阵，并将混淆模式与Filterbank和Wav2vec 2.0系统进行了比较。我们的原始波形声学模型由参数化（Sinc2Net）或非参数化CNN以及双向LSTM组成，在TIMIT开发/测试集上实现了低至13.7%/15.2%的PER，优于文献中报道的原始波形模型的PER。我们还研究了从WSJ进行迁移学习对语音错误模式和混淆矩阵的影响。迁移学习将开发/测试集上的PER降低至11.8%/13.7%。