用神经网络集成生成语音边界置信区间，更真实反映发音过渡与模型不确定性。
Gradient boundaries through confidence intervals for forced alignment estimates using model ensembles
- 通过10个神经网络模型的集成，以中位数确定边界点。
- 使用97.85%置信区间构建梯度边界范围，体现模型置信度。
- 结果可输出为JSON和Praat TextGrid，适合语音分析与人工校验。
强制对齐是将音频与音素或拼写转录对齐的常用工具。现有方法通常仅提供边界点估计。本文提出一种基于神经网络集成的方法，通过多次运行不同分类器获得边界分布，以中位数作为点估计，利用顺序统计量构建97.85%置信区间得到梯度边界范围。该方法更真实地刻画了语音段之间的过渡，并显式表达模型对边界位置的不确定性，有助于识别需人工审查的边界。在Buckeye和TIMIT数据集上，集成方法整体性能略优于单模型。边界信息可输出为JSON文件和主表格，便于程序化分析；同时兼容Praat TextGrid格式，以点层表示边界区域。
原文摘要 · Abstract (English)
Forced alignment is a common tool to align audio with orthographic and phonetic transcriptions. Most forced alignment tools provide only point-estimates of boundaries. The present project introduces a method of producing gradient boundaries by deriving confidence intervals using neural network ensembles. Ten different segment classifier neural networks were previously trained, and the alignment process is repeated with each classifier. The ensemble is then used to place the point-estimate of a boundary at the median of the boundaries in the ensemble, and the gradient range is placed using a 97.85% confidence interval around the median constructed using order statistics. Gradient boundaries are taken here as a more realistic representation of how segments transition into each other. Moreover, the range indicates the model uncertainty in the boundary placement, facilitating tasks like finding boundaries that should be reviewed. As a bonus, on the Buckeye and TIMIT corpora, the ensemble boundaries show a slight overall improvement over using just a single model. The gradient boundaries can be emitted during alignment as JSON files and a main table for programmatic and statistical analysis. For familiarity, they are also output as Praat TextGrids using a point tier to represent the edges of the boundary regions.
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。