Cover Image

Lost in Transcription, Found in Distribution Shift: Demystifying Hallucination in Speech Foundation Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Atwany, Hanin, Waheed, Abdul, Singh, Rita, Choudhury, Monojit, Raj, Bhiksha
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

What Do Speech Foundation Models Not Learn About Speech?
by: Waheed, Abdul, et al.
Published: (2024)

On the Robust Approximation of ASR Metrics
by: Waheed, Abdul, et al.
Published: (2025)

DELULU: Discriminative Embedding Learning Using Latent Units for Speaker-Aware Self-Trained Speech Foundational Model
by: Baali, Massa, et al.
Published: (2025)

Speech vs. Transcript: Does It Matter for Human Annotators in Speech Summarization?
by: Sharma, Roshan, et al.
Published: (2024)

uDistil-Whisper: Label-Free Data Filtering for Knowledge Distillation in Low-Data Regimes
by: Waheed, Abdul, et al.
Published: (2024)

VideoJudge: Bootstrapping Enables Scalable Supervision of MLLM-as-a-Judge for Video Understanding
by: Waheed, Abdul, et al.
Published: (2025)

Less is More Tokens: Efficient Math Reasoning via Difficulty-Aware Chain-of-Thought Distillation
by: Waheed, Abdul, et al.
Published: (2025)

What and When to Learn: CURriculum Ranking Loss for Large-Scale Speaker Verification
by: Baali, Massa, et al.
Published: (2026)

On the Diversity of Synthetic Data and its Impact on Training Large Language Models
by: Chen, Hao, et al.
Published: (2024)

User Behavior Prediction as a Generic, Robust, Scalable, and Low-Cost Evaluation Strategy for Estimating Generalization in LLMs
by: Saha, Sougata, et al.
Published: (2025)

CAARMA: Class Augmentation with Adversarial Mixup Regularization
by: Baali, Massa, et al.
Published: (2025)

PDAF: A Phonetic Debiasing Attention Framework For Speaker Verification
by: Baali, Massa, et al.
Published: (2024)

CoLMbo: Speaker Language Model for Descriptive Profiling
by: Baali, Massa, et al.
Published: (2025)

OleSpeech-IV: A Large-Scale Multispeaker and Multilingual Conversational Speech Dataset with Diverse Topics
by: Chu, Wei, et al.
Published: (2025)

Token Prediction as Implicit Classification to Identify LLM-Generated Text
by: Chen, Yutian, et al.
Published: (2023)

[WIP] Jailbreak Paradox: The Achilles' Heel of LLMs
by: Rao, Abhinav, et al.
Published: (2024)

No Encore: Unlearning as Opt-Out in Music Generation
by: Kim, Jinju, et al.
Published: (2025)

Evaluating Large Language Models for Health-related Queries with Presuppositions
by: Kaur, Navreet, et al.
Published: (2023)

SVeritas: Benchmark for Robust Speaker Verification under Diverse Conditions
by: Baali, Massa, et al.
Published: (2025)

Hallucination Benchmark for Speech Foundation Models
by: Koudounas, Alkis, et al.
Published: (2025)

Towards Zero-Shot Text-To-Speech for Arabic Dialects
by: Doan, Khai Duy, et al.
Published: (2024)

To Generate or Discriminate? Methodological Considerations for Measuring Cultural Alignment in LLMs
by: Pandey, Saurabh Kumar, et al.
Published: (2026)

Lost in Transcription: Identifying and Quantifying the Accuracy Biases of Automatic Speech Recognition Systems Against Disfluent Speech
by: Mujtaba, Dena, et al.
Published: (2024)

Mark My Words: A Robust Multilingual Model for Punctuation in Text and Speech Transcripts
by: Pulipaka, Sidharth, et al.
Published: (2025)

Demystifying ChatGPT: How It Masters Genre Recognition
by: Raj, Subham, et al.
Published: (2025)

DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
by: Li, Zherui, et al.
Published: (2025)

Meta-Cultural Competence: Climbing the Right Hill of Cultural Awareness
by: Saha, Sougata, et al.
Published: (2025)

PhoniTale: Phonologically Grounded Mnemonic Generation for Typologically Distant Language Pairs
by: Kang, Sana, et al.
Published: (2025)

Litmus (Re)Agent: A Benchmark and Agentic System for Predictive Evaluation of Multilingual Models
by: Mittal, Avni, et al.
Published: (2026)

Revisiting Acoustic Features for Robust ASR
by: Shah, Muhammad A., et al.
Published: (2024)

Hallucinations in Neural Automatic Speech Recognition: Identifying Errors and Hallucinatory Models
by: Frieske, Rita, et al.
Published: (2024)

Sacred or Synthetic? Evaluating LLM Reliability and Abstention for Religious Questions
by: Atif, Farah, et al.
Published: (2025)

WhisperRT -- Turning Whisper into a Causal Streaming Model
by: Krichli, Tomer, et al.
Published: (2025)

Lost in Phonation: Voice Quality Variation as an Evaluation Dimension for Speech Foundation Models
by: Lameris, Harm, et al.
Published: (2025)

Do Moral Judgment and Reasoning Capability of LLMs Change with Language? A Study using the Multilingual Defining Issues Test
by: Khandelwal, Aditi, et al.
Published: (2024)

Ethical Reasoning and Moral Value Alignment of LLMs Depend on the Language we Prompt them in
by: Agarwal, Utkarsh, et al.
Published: (2024)

Speech Robust Bench: A Robustness Benchmark For Speech Recognition
by: Shah, Muhammad A., et al.
Published: (2024)

Women, Infamous, and Exotic Beings: A Comparative Study of Honorific Usages in Wikipedia and LLMs for Bengali and Hindi
by: Mukherjee, Sourabrata, et al.
Published: (2025)

Tricking LLMs into Disobedience: Formalizing, Analyzing, and Detecting Jailbreaks
by: Rao, Abhinav, et al.
Published: (2023)

Impact of Noisy Supervision in Foundation Model Learning
by: Chen, Hao, et al.
Published: (2024)