Quantifying Bias in Automatic Speech Recognition

2021-03-28 · Siyuan Feng, Olya Kudina, Bence Mark Halpern, Odette Scharenborg

Automatic speech recognition (ASR) systems promise to deliver objective interpretation of human speech. Practice and recent evidence suggests that the state-of-the-art (SotA) ASRs struggle with the large variation in speech due to e.g., gender, age, speech impairment, race, and accents. Many factors can cause the bias of an ASR system. Our overarching goal is to uncover bias in ASR systems to work towards proactive bias mitigation in ASR. This paper is a first step towards this goal and systematically quantifies the bias of a Dutch SotA ASR system against gender, age, regional accents and non-native accents. Word error rates are compared, and an in-depth phoneme-level error analysis is conducted to understand where bias is occurring. We primarily focus on bias due to articulation differences in the dataset. Based on our findings, we suggest bias mitigation strategies for ASR development.

📄 PDF Abstract BibTeX arXiv:2103.15122

Code (1)

syfengcuhk/jasmin 공식 구현

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognition Speech Recognition

How to Evaluate Automatic Speech Recognition: Comparing Different Performance and Bias Measures

2025-07-08 · Tanvina Patel, Wiebke Hutiri, Aaron Yi Ding, Odette Scharenborg arxiv

There is increasingly more evidence that automatic speech recognition (ASR) systems are biased against different speakers and speaker groups, e.g., due to gender, age, or accent. Research on bias in ASR has so far primar…

Speech Recognition

Lost in Transcription: Identifying and Quantifying the Accuracy Biases of Automatic Speech Recognition Systems Against Disfluent Speech

2024-05-10 · Dena Mujtaba, Nihar R. Mahapatra, Megan Arney, J. Scott Yaruss 외

Automatic speech recognition (ASR) systems, increasingly prevalent in education, healthcare, employment, and mobile technology, face significant challenges in inclusivity, particularly for the 80 million-strong global co…

Automatic Speech Recognition Automatic Speech Recognition (ASR)speech-recognition Speech Recognition

The Balancing Act: Unmasking and Alleviating ASR Biases in Portuguese

2024-02-12 · Ajinkya Kulkarni, Anna Tokareva, Rameez Qureshi, Miguel Couceiro

In the field of spoken language understanding, systems like Whisper and Multilingual Massive Speech (MMS) have shown state-of-the-art performances. This study is dedicated to a comprehensive exploration of the Whisper an…

Automatic Speech Recognition Automatic Speech Recognition (ASR)speech-recognition Speech Recognition+1

OWSM-Biasing: Contextualizing Open Whisper-Style Speech Models for Automatic Speech Recognition with Dynamic Vocabulary

2025-06-11 · Yui Sudo, Yusuke Fujita, Atsushi Kojima, Tomoya Mizumoto 외

Speech foundation models (SFMs), such as Open Whisper-Style Speech Models (OWSM), are trained on massive datasets to achieve accurate automatic speech recognition. However, even SFMs struggle to accurately recognize rare…

Automatic Speech Recognition speech-recognition Speech Recognition

Debiased Automatic Speech Recognition for Dysarthric Speech via Sample Reweighting with Sample Affinity Test

2023-05-22 · Eungbeom Kim, Yunkee Chae, Jaeheon Sim, Kyogu Lee

Automatic speech recognition systems based on deep learning are mainly trained under empirical risk minimization (ERM). Since ERM utilizes the averaged performance on the data samples regardless of a group such as health…