SpeechLLMs
PulseAugur coverage of SpeechLLMs — every cluster mentioning SpeechLLMs across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
SpeechLLM decoders show significant redundancy, allowing for layer pruning
Researchers have investigated the redundancy within decoder layers of Speech Large Language Models (SpeechLLMs), which typically comprise over 90% of the model's parameters. Their study across various model sizes reveal…
-
FBK's SpeechLLMs achieve strong results in IWSLT 2026 instruction following task
FBK researchers have developed SpeechLLMs for the IWSLT 2026 Instruction Following shared task, focusing on both short-form and long-form speech instruction following. For short-form tasks, their model achieved a SIFS s…
-
SpeechLLMs show bias against Eastern European accents, research finds
A new research paper has quantified intersectional bias in Speech Large Language Models (SpeechLLMs). The study used 2,880 controlled interactions across six English accents and two gender presentations, employing voice…
-
New methods enhance simultaneous speech translation with decoder-only LLMs
Researchers are developing new methods for simultaneous speech translation, focusing on decoder-only large language models. One approach, AlignAtt4LLM, adapts attention mechanisms for these models to improve translation…
-
New Research Exposes Privacy Risks in Domain-Adapted SpeechLLMs
A new research paper published on arXiv details a significant privacy risk in domain-adapted Automatic Speech Recognition (ASR) models, often referred to as SpeechLLMs. The study reveals that when these models are custo…
-
SpeechLLMs show mixed results in translation benchmarks, cascades still lead
A new comprehensive test suite, Hearing to Translate, has been developed to evaluate the effectiveness of integrating speech modality directly into Large Language Models (LLMs) for speech-to-text translation. The study …