Pith. sign in

REVIEW 5 cited by

Quantifying Bias in Automatic Speech Recognition

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2103.15122 v2 pith:OV73LV5U submitted 2021-03-28 eess.AS cs.CLcs.SD

classification eess.AScs.CLcs.SD
keywords biasspeechaccentsautomaticerrorgendergoalmitigation
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 53 citations worldwide. Full citation record

  1. Automatic Speech Recognition Biases in Newcastle English: an Error Analysis

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Automatic speech recognition errors on Newcastle English are driven primarily by regional dialect features such as glottalisation, monophthongisation, and local pronouns, rather than by speaker age or gender.

  2. Addressing Pitfalls in Auditing Practices of Automatic Speech Recognition Technologies: A Case Study of People with Aphasia

    cs.CY 2025-06 conditional novelty 6.0 of 10

    Across six ASR services, speakers with aphasia receive worse transcriptions than controls, and standard audit methods mask within-group disparities and hallucination risks.

  3. ASKD-Whisper: Adaptive Self-knowledge Distillation for Efficient and Low-Latency Automatic Speech Recognition

    cs.CL 2026-01 reject novelty 4.0 of 10

    A distillation method that decays teacher loss then applies self-distillation yields a Whisper-derived ASR model with 5x lower latency and slightly better average WER only on in-domain noisy datasets.

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

    cs.CL 2025-07 conditional novelty 4.0 of 10

    On Dutch end-to-end ASR systems, average word error rate hides large performance gaps across speaker groups, and bias mitigation can lower the average while increasing bias.

  5. FairASR: Fair Audio Contrastive Learning for Automatic Speech Recognition

    eess.AS 2025-06 conditional novelty 4.0 of 10

    FairASR pretrains a Conformer with InfoNCE plus a gradient-reversed supervised contrastive loss over demographic labels, reducing demographic WER gaps on FairSpeech with small overall WER cost.

Pith tools