BaLEEN (Biasing with Latent Encoded Entities), a lightweight, hypernetwork-based framework for dynamic contextual adaptation without fine-tuning the underlying ASR model, is proposed.
Abstract
Transcribing domain-specific entities and rare proper nouns remains a major challenge in automatic speech recognition (ASR). In this paper, we propose BaLEEN (Biasing with Latent Encoded Entities), a lightweight, hypernetwork-based framework for dynamic contextual adaptation without fine-tuning the underlying ASR model. BaLEEN encodes variable-length contextual keywords using a pretrained language model, compresses them into a fixed sequence of latent vectors via a Perceiver bottleneck, and injects context-dependent bias vectors directly into the intermediate encoder representations of the ASR model. Because both the language model and the backbone ASR model remain entirely frozen during training, BaLEEN operates as a plug-and-play adapter that incurs zero computational overhead at inference time when context biases are precomputed. We evaluate our method on a CTC-based ASR model using a Wikipedia-derived corpus with annotated named entities and synthetic speech. Experimental results demonstrate that BaLEEN reduces keyword miss rate by 8.7% on the test set relative to the unbiased baseline while simultaneously improving overall word error rate by 21% and character error rate by 28%.
This work curates a human-annotated topic-utterance judgments dataset and examines the superior performance of lightweight LLM matchers over embedding and regex models when equipped with natural language descriptions.
Saman Rahbar, Xi-Liang Zhu, Irvin Cardoza et al.· 0 citations
This paper describes the system for Task~2 of the second Multilingual Conversational Speech Language Model (MLC-SLM) Challenge, which adapt Qwen3-Omni-30B-A3B-Instruct with a segment-evidence-aware data and post-training pipeline and obtains 90.92% accuracy on the final official evaluation set.
H. Le, L. Nguyen, Minh Tri Dao· 2nd Workshop on Multilingual...· 1 citation
In-context learning (ICL) can be amortized into latent objects (task vectors, function vectors, context vectors) that recover few-shot behavior at zero-shot inference cost, but recent theory shows a static vector acts as a single synthetic demonstration and must fail on high-rank mappings such as word-level bijections....
Standard Automatic Speech Recognition (ASR) systems struggle to accurately transcribe dense, domain-specific terminology in academic lectures. This paper introduces a dual-stage context injection pipeline that enhances transcription quality through a multi-stage Retrieval-Augmented Generation (RAG) framework. Domain kn...
Andreas Lindner· Automation, Control, and Inf...· 0 citations
To improve span localization, ASR encoder states are reuse ASR encoder states as sentence-level representations and fuse them with textual embeddings to let lightweight span locators exploit speech information without running a separate audio encoder.
Steffen Freisinger, Philipp Seeberger, Thomas Ranzenberger et al.· 0 citations
A linguistically curated bilingual dataset comprising 13,846 utterances distributed across synthetic and literature-derived subsets, with OOV rates reaching up to 100%, and a diagnostic evaluation framework that partitions recognition performance into Biased Word Error Rate (B-WER), which targets domain-specific jargon...
Emily Haubert Klering, E. Cortes, Т. И. Черненко et al.· International Conference on...· 0 citations