Cover Image

Conversational Speech Reveals Structural Robustness Failures in SpeechLLM Backbones

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Teleki, Maria, Janjur, Sai, Liu, Haoran, Grabner, Oliver, Verma, Ketan, Docog, Thomas, Dong, Xiangjue, Shi, Lingfeng, Wang, Cong, Birkelbach, Stephanie, Kim, Jason, Zhang, Yin, Székely, Éva, Caverlee, James
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence Audio and Speech Processing
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Z-Scores: A Metric for Linguistically Assessing Disfluency Removal
by: Teleki, Maria, et al.
Published: (2025)

The Voice Behind the Words: Quantifying Intersectional Bias in SpeechLLMs
by: Satish, Shree Harsha Bokkahalli, et al.
Published: (2026)

When Voice Matters: Evidence of Gender Disparity in Positional Bias of SpeechLLMs
by: Satish, Shree Harsha Bokkahalli, et al.
Published: (2025)

Do Bias Benchmarks Generalise? Evidence from Voice-based Evaluation of Gender Bias in SpeechLLMs
by: Satish, Shree Harsha Bokkahalli, et al.
Published: (2025)

Evaluating Text-to-Speech Synthesis from a Large Discrete Token-based Speech Language Model
by: Wang, Siyang, et al.
Published: (2024)

SpeechLLM-as-Judges: Towards General and Interpretable Speech Quality Evaluation
by: Wang, Hui, et al.
Published: (2025)

Streaming Speech-to-Text Translation with a SpeechLLM
by: Parcollet, Titouan, et al.
Published: (2026)

S2ST-Omni: Hierarchical Language-Aware SpeechLLM Adaptation for Multilingual Speech-to-Speech Translation
by: Pan, Yu, et al.
Published: (2025)

SocialPulse: An Open-Source Subreddit Sensemaking Toolkit
by: Birkelbach, Stephanie, et al.
Published: (2026)

Masculine Defaults via Gendered Discourse in Podcasts and Large Language Models
by: Teleki, Maria, et al.
Published: (2025)

What Counts as Real? Speech Restoration and Voice Quality Conversion Pose New Challenges to Deepfake Detection
by: Satish, Shree Harsha Bokkahalli, et al.
Published: (2026)

A Survey on LLM Inference-Time Self-Improvement
by: Dong, Xiangjue, et al.
Published: (2024)

Better Pseudo-labeling with Multi-ASR Fusion and Error Correction by SpeechLLM
by: Prakash, Jeena, et al.
Published: (2025)

Qwen vs. Gemma Integration with Whisper: A Comparative Study in Multilingual SpeechLLM Systems
by: Nguyen, Tuan, et al.
Published: (2025)

Rubric-Guided Fine-tuning of SpeechLLMs for Multi-Aspect, Multi-Rater L2 Reading-Speech Assessment
by: Parikh, Aditya Kamlesh, et al.
Published: (2026)

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

MamTra: A Hybrid Mamba-Transformer Backbone for Speech Synthesis
by: Nguyen, Tan Dat, et al.
Published: (2026)

Conversational Speech Naturalness Predictor
by: Xu, Anfeng, et al.
Published: (2026)

VC-ENHANCE: Speech Restoration with Integrated Noise Suppression and Voice Conversion
by: Byun, Kyungguen, et al.
Published: (2024)

Speak Your Mind: The Speech Continuation Task as a Probe of Voice-Based Model Bias
by: Satish, Shree Harsha Bokkahalli, et al.
Published: (2025)

CHOIR: Collaborative Harmonization fOr Inference Robustness
by: Dong, Xiangjue, et al.
Published: (2025)

DisfluencySpeech -- Single-Speaker Conversational Speech Dataset with Paralanguage
by: Wang, Kyra, et al.
Published: (2024)

Voice-ENHANCE: Speech Restoration using a Diffusion-based Voice Conversion Framework
by: Byun, Kyungguen, et al.
Published: (2025)

UniTalker: Conversational Speech-Visual Synthesis
by: Hu, Yifan, et al.
Published: (2025)

Attempt Towards Stress Transfer in Speech-to-Speech Machine Translation
by: Akarsh, Sai, et al.
Published: (2024)

Switchboard-Affect: Emotion Perception Labels from Conversational Speech
by: Romana, Amrit, et al.
Published: (2025)

Conformer-based Ultrasound-to-Speech Conversion
by: Ibrahimov, Ibrahim, et al.
Published: (2025)

Benchmarking Humans and Machines on Complex Multilingual Speech Understanding Tasks
by: Kankanala, Sai Samrat, et al.
Published: (2025)

The TEA-ASLP System for Multilingual Conversational Speech Recognition and Speech Diarization in MLC-SLM 2025 Challenge
by: Xue, Hongfei, et al.
Published: (2025)

Advancing Speech Translation: A Corpus of Mandarin-English Conversational Telephone Speech
by: Wotherspoon, Shannon, et al.
Published: (2024)

Generating Data with Text-to-Speech and Large-Language Models for Conversational Speech Recognition
by: Cornell, Samuele, et al.
Published: (2024)

Generative Expressive Conversational Speech Synthesis
by: Liu, Rui, et al.
Published: (2024)

Enhancing In-the-Wild Speech Emotion Conversion with Resynthesis-based Duration Modeling
by: Prabhu, Navin Raj, et al.
Published: (2025)

Benchmarking Speech Systems for Frontline Health Conversations: The DISPLACE-M Challenge
by: E, Dhanya, et al.
Published: (2026)

Textless Streaming Speech-to-Speech Translation using Semantic Speech Tokens
by: Zhao, Jinzheng, et al.
Published: (2024)

Pheme: Efficient and Conversational Speech Generation
by: Budzianowski, Paweł, et al.
Published: (2024)

AS-Speech: Adaptive Style For Speech Synthesis
by: Li, Zhipeng, et al.
Published: (2024)

Adapting Whisper for Streaming Speech Recognition via Two-Pass Decoding
by: Zhou, Haoran, et al.
Published: (2025)

RAVE for Speech: Efficient Voice Conversion at High Sampling Rates
by: Bargum, Anders R., et al.
Published: (2024)

A Unified Speech LLM for Diarization and Speech Recognition in Multilingual Conversations
by: Saengthong, Phurich, et al.
Published: (2025)