paper-with-me

홈 › Papers

Beyond WER: Probing Whisper's Sub-token Decoder Across Diverse Language Resource Levels

2025-09-29 · Siyu Liang, Nicolas Ballier, Gina-Anne Levow, Richard Wright arxiv

While large multilingual automatic speech recognition (ASR) models achieve remarkable performance, the internal mechanisms of the end-to-end pipeline, particularly concerning fairness and efficacy across languages, remain underexplored. This paper introduces a fine-grained analysis of Whisper's multilingual decoder, examining its sub-token hypotheses during transcription across languages with various resource levels. Our method traces the beam search path, capturing sub-token guesses and their associated probabilities. Results reveal that higher resource languages benefit from higher likelihood of the correct token being top-ranked, greater confidence, lower predictive entropy, and more diverse alternative candidates. Lower resource languages fare worse on these metrics, but also exhibit distinct clustering patterns in sub-token usage sometimes influenced by typology in our PCA and t-SNE analysis. This sub-token probing uncovers systematic decoding disparities masked by aggregate error rates and points towards targeted interventions to ameliorate the imbalanced development of speech technology.

📄 PDF Abstract BibTeX arXiv:2509.25516

Code (0)

등록된 구현이 없습니다.

Tasks

Speech Recognition

Similar Papers 제목 키워드 기반

From Text Metrics to Model Internals: A Study of Whisper ASR Hallucination Detection

2026-06-22 · Jan Jasiński, Mateusz Barański, Julitta Bartolewska, Marcin Witkowski 외 arxiv

Hallucinations of ASR models - fluent transcriptions with no basis in audio - degrade system performance and pose risks in downstream applications. Robust detection of such errors remains a challenge. This paper studies …

Whisper-CD: Accurate Long-Form Speech Recognition using Multi-Negative Contrastive Decoding

2026-03-06 · Hoseong Ahn, Jeongyun Chae, Yoonji Park, Kyuhong Shim arxiv

Long-form speech recognition with large encoder-decoder models such as Whisper often exhibit hallucinations, repetition loops, and content omissions. These errors can accumulate and be further amplified when the previous…

Speech Recognition

WhisperRT -- Turning Whisper into a Causal Streaming Model

2025-08-17 · Tomer Krichli, Bhiksha Raj, Joseph Keshet arxiv

Automatic Speech Recognition (ASR) has seen remarkable progress, with models like OpenAI Whisper and NVIDIA Canary achieving state-of-the-art (SOTA) performance in offline transcription. However, these models are not des…

Speech Recognition

Overcoming Decoder Inconsistencies in Whisper for Dravidian and Low-Resource Languages

2026-06-08 · Chowdam Venkata Kumar, Kumud Tripathi, Pankaj Wasnik arxiv

Multilingual ASR models such as Whisper perform well on high-resource languages but exhibit substantially higher Word Error Rates (WER) for Dravidian languages compared to Indo-Aryan ones. Through linguistic and dataset …

Keyword-Guided Adaptation of Automatic Speech Recognition

2024-06-04 · Aviv Shamsian, Aviv Navon, Neta Glazer, Gill Hetz 외

Automatic Speech Recognition (ASR) technology has made significant progress in recent years, providing accurate transcription across various domains. However, some challenges remain, especially in noisy environments and …

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)DecoderKeyword Spotting+2