LaSR: Context-Aware Speech Recognition via Latent Reasoning

• URL: http://arxiv.org/abs/2606.00507v1
• Date: Sat, 30 May 2026 03:44:12 GMT
• Title: LaSR: Context-Aware Speech Recognition via Latent Reasoning
• Abstract summary: We propose LaSR (Latent Speech Reasoning), a novel training paradigm featuring a context-aware reasoning trajectory.<n>Our findings highlight the potential of latent reasoning in building efficient, context-aware speech assistants.
• Score: 29.32686483648477
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in Speech Large Language Models (Speech LLMs) have significantly enhanced spoken language understanding and reasoning. However, their contextual awareness is limited, struggling to perform speech recognition that effectively reflects the speaker's intent and topical context. In this paper, we propose LaSR (Latent Speech Reasoning), a novel training paradigm featuring a context-aware reasoning trajectory that leverages the latent reasoning process. Instead of generating explicit intermediate tokens, LaSR aligns chain-of-thought (CoT) supervision around the acoustic feature region of the targeted word, and introduces latent reasoning periods for context information grounding and transcriptional transition. Furthermore, to effectively benchmark contextual recognition on specialized vocabulary, we propose Spoken Darwin-Science, a large-scale corpus focusing on academic terminologies. Preliminary experiments on Fun-Audio-Chat demonstrate that LaSR significantly improves terminology recognition without introducing additional latency and consistently outperforms standard supervised fine-tuning baselines. Our findings highlight the potential of latent reasoning in building efficient, context-aware speech assistants.

Related papers

• Bridging the Gap: Converting Read Text to Conversational Dialogue [1.3553462938197818]
  This paper introduces a novel approach, Prosodic Adjustment with Conversational Context (PACC)<n>PACC converts read speech into natural conversational speech used in various modern applications.<n>Unlike conventional methods, our approach uses High-Fidelity Generative Adversarial Networks (HiFi-GAN) for speech synthesis.
  arXiv  Detail & Related papers  (2026-05-18T07:53:46Z)
• The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning [76.96796481909581]
  Internal cognitive processing is instrumental in formulating high-quality responses.<n>We propose a novel method named FLAIR that conducts latent thinking simultaneously with speech perception.<n>Our approach aligns seamlessly with spoken dialogue systems.
  arXiv  Detail & Related papers  (2026-03-18T15:30:29Z)
• VowelPrompt: Hearing Speech Emotions from Text via Vowel-level Prosodic Augmentation [34.905479321921575]
  We propose VowelPrompt, a framework that augments large language models with interpretable, fine-grained vowel-level prosodic cues.<n>We show that VowelPrompt consistently outperforms state-of-the-art emotion recognition methods under zero-shot, fine-tuned, cross-domain, and cross-linguistic conditions.
  arXiv  Detail & Related papers  (2026-02-06T00:09:14Z)
• MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance [66.74042564585942]
  MOSS-Speech is a true speech-to-speech large language model that directly understands and generates speech without relying on text guidance.<n>Our work establishes a new paradigm for expressive and efficient end-to-end speech interaction.
  arXiv  Detail & Related papers  (2025-10-01T04:32:37Z)
• Cross-Modal Knowledge Distillation for Speech Large Language Models [10.840179376551804]
  We show that introducing speech capabilities can degrade knowledge and reasoning even when inputs remain textual.<n>We propose a cross-modal knowledge distillation framework that leverages both text-to-text and speech-to-text channels to transfer knowledge from a text-based teacher model to a speech LLM.
  arXiv  Detail & Related papers  (2025-09-18T13:07:53Z)
• Mini-Omni-Reasoner: Token-Level Thinking-in-Speaking in Large Speech Models [80.75260664100644]
  Mini-Omni-Reasoner is a framework that enables reasoning within speech via a novel "Thinking-in-Speaking" formulation.<n>It interleaves silent reasoning tokens with spoken response tokens at the token level.<n>It achieves a +19.1% gain in arithmetic reasoning and +6.4% in contextual understanding, with shorter outputs and zero decoding latency.
  arXiv  Detail & Related papers  (2025-08-18T15:14:04Z)
• SpeechR: A Benchmark for Speech Reasoning in Large Audio-Language Models [60.72029578488467]
  SpeechR is a unified benchmark for evaluating reasoning over speech in large audio-language models.<n>It evaluates models along three key dimensions: factual retrieval, procedural inference, and normative judgment.<n> Evaluations on eleven state-of-the-art LALMs reveal that high transcription accuracy does not translate into strong reasoning capabilities.
  arXiv  Detail & Related papers  (2025-08-04T03:28:04Z)
• ProsodyLM: Uncovering the Emerging Prosody Processing Capabilities in Speech Language Models [70.56468982313834]
  We propose ProsodyLM, which introduces a simple tokenization scheme amenable to learning prosody.<n>We find that ProsodyLM can learn surprisingly diverse emerging prosody processing capabilities through pre-training alone.
  arXiv  Detail & Related papers  (2025-07-27T00:59:01Z)
• Speech-IFEval: Evaluating Instruction-Following and Quantifying Catastrophic Forgetting in Speech-Aware Language Models [49.1574468325115]
  We introduce Speech-IFeval, an evaluation framework designed to assess instruction-following capabilities.<n>Recent SLMs integrate speech perception with large language models (LLMs), often degrading textual capabilities due to speech-centric training.<n>Our findings show that most SLMs struggle with even basic instructions, performing far worse than text-based LLMs.
  arXiv  Detail & Related papers  (2025-05-25T08:37:55Z)
• Internalizing ASR with Implicit Chain of Thought for Efficient Speech-to-Speech Conversational LLM [3.6950912517562435]
  We propose a method that implicitly internalizes ASR chain of thought into a speech LLM, enhancing its native speech understanding capabilities. Our approach reduces latency and improves the model's native understanding of speech, paving the way for more efficient and natural real-time audio interactions.
  arXiv  Detail & Related papers  (2024-09-25T20:59:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.