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.

在这篇论文中，我们分析了 TIMIT 语音识别任务中原始波形声学模型的错误模式，对传统的音素错误率指标进行了超越。我们将音素分为三组，计算了每个广义音素类别的音素错误率，并构建了每个类别的混淆矩阵，与 Filterbank 和 Wav2vec 2.0 系统的混淆模式进行了比较。我们的原始波形声学模型采用参数化（Sinc2Net）或非参数化 CNNs 和双向 LSTMs，并在 TIMIT 开发 / 测试集上实现了 13.7%/15.2% 的音素错误率，超过了文献中报告的原始波形模型的音素错误率。我们还研究了从 WSJ 进行的迁移学习对音素错误模式和混淆矩阵的影响，将音素错误率降低到了 11.8%/13.7%。