Pseudo2Real：自动语音识别中伪标签校正的任务算术方法

摘要

在领域转移下实现稳健的自动语音识别（ASR）至关重要，因为现实世界中的系统常会遇到未见过的口音和领域，且标注数据有限。尽管伪标签提供了一种实用的解决方案，但它往往会引入系统性的、特定于口音的误差，这些误差通过过滤难以消除。我们提出疑问：在没有目标领域真实标签的情况下，如何纠正这些重复出现的偏差？我们提出了一种简单的参数空间校正方法：在包含真实标签和伪标签数据的源域中，从相同的初始化状态微调两个ASR模型，一个使用真实标签，另一个使用伪标签，它们之间的权重差异形成校正向量，该向量捕捉了伪标签的偏差。当将此向量应用于伪标签的目标模型时，它提升了识别效果，在Whisper tiny模型上对AfriSpeech-200数据集中的十种非洲口音测试，实现了高达35%的相对词错误率（WER）降低。

English

Robust ASR under domain shift is crucial because real-world systems encounter unseen accents and domains with limited labeled data. Although pseudo-labeling offers a practical workaround, it often introduces systematic, accent-specific errors that filtering fails to fix. We ask: How can we correct these recurring biases without target ground truth? We propose a simple parameter-space correction: in a source domain containing both real and pseudo-labeled data, two ASR models are fine-tuned from the same initialization, one on ground-truth labels and the other on pseudo-labels, and their weight difference forms a correction vector that captures pseudo-label biases. When applied to a pseudo-labeled target model, this vector enhances recognition, achieving up to a 35% relative Word Error Rate (WER) reduction on AfriSpeech-200 across ten African accents with the Whisper tiny model.

Pseudo2Real：自动语音识别中伪标签校正的任务算术方法

Pseudo2Real: Task Arithmetic for Pseudo-Label Correction in Automatic Speech Recognition

摘要

Support