# Think Before You Transcribe: Context-Steered Reasoning for Hotword-Aware Speech Recognition

arXiv: https://arxiv.org/abs/2610.26986 · HTML: https://mlsanity.com/paper/2610.26986/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Speech recognition models and training, one of 11 directions in Speech and audio language models.
- Builds on 18 of its 29 references that are in the map.
- First posted on arXiv 19 Oct 2026.

## Placed in

- Speech and audio language models / [Speech recognition models and training](https://mlsanity.com/speech/speech-recognition-models-and-training/)

## Builds on (most cited first, 18 of 18)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [LLaVA-CoT: Let Vision Language Models Reason Step-by-Step](https://mlsanity.com/paper/2411.10440/) (2024)
- [Baichuan-Omni-1.5 Technical Report](https://mlsanity.com/paper/2501.15368/) (2025)
- [Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction](https://mlsanity.com/paper/2502.11946/) (2025)
- [Audio-CoT: Exploring Chain-of-Thought Reasoning in Large Audio Language Model](https://mlsanity.com/paper/2501.07246/) (2025)
- [MiDashengLM: Efficient Audio Understanding with General Audio Captions](https://mlsanity.com/paper/2508.03983/) (2025)
- [SlideSpeech: A Large-Scale Slide-Enriched Audio-Visual Corpus](https://mlsanity.com/paper/2309.05396/) (2023)
- [Mind with Eyes: from Language Reasoning to Multimodal Reasoning](https://mlsanity.com/paper/2503.18071/) (2025)
- [ContextASR-Bench: A Massive Contextual Speech Recognition Benchmark](https://mlsanity.com/paper/2507.05727/) (2025)
- [CopyNE: Better Contextual ASR by Copying Named Entities](https://mlsanity.com/paper/2305.12839/) (2023)
- [Contextualization of ASR with LLM using phonetic retrieval-based augmentation](https://mlsanity.com/paper/2409.15353/) (2024)
- [Group Relative Policy Optimization for Speech Recognition](https://mlsanity.com/paper/2509.01939/) (2025)
- [DeRAGEC: Denoising Named Entity Candidates with Synthetic Rationale for ASR Error Correction](https://mlsanity.com/paper/2506.07510/) (2025)
- [OWSM-Biasing: Contextualizing Open Whisper-Style Speech Models for Automatic Speech Recognition with Dynamic Vocabulary](https://mlsanity.com/paper/2506.09448/) (2025)
- [Speech LLMs are Contextual Reasoning Transcribers](https://mlsanity.com/paper/2604.00610/) (2026)
- [Valley3: Scaling Omni Foundation Models for E-commerce](https://mlsanity.com/paper/2605.01278/) (2026)

## Abstract (arXiv)

Contextual automatic speech recognition (ASR) improves rare-entity recognition by supplying candidate hotwords, but retrieved candidates may be incomplete or incorrect. A reliable system must therefore exploit a relevant candidate without allowing unsupported alternatives to override the acoustic evidence. We propose a context-steered reasoning framework that generates a short, audio-grounded caption before transcription. This task-specific representation lets the model assess candidate relevance before making a lexical decision. Supervised fine-tuning establishes the structured caption-to-transcription behavior, while targeted trajectory injection strengthens group relative policy optimization when difficult utterances yield only unsuccessful rollouts. Experiments across diverse context conditions show that the proposed method consistently improves both transcription quality and hotword recognition while remaining robust to irrelevant candidates.