# Closing the Gap Between Text and Speech Understanding in LLMs

arXiv: https://arxiv.org/abs/2510.13632 · HTML: https://mlsanity.com/paper/2510.13632/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 44 references that are in the map.
- First posted on arXiv 15 Oct 2025.

## Builds on (most cited first, 29 of 29)

- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [SALMONN: Towards Generic Hearing Abilities for Large Language Models](https://mlsanity.com/paper/2310.13289/) (2023)
- [TinyStories: How Small Can Language Models Be and Still Speak Coherent English?](https://mlsanity.com/paper/2305.07759/) (2023)
- [Scaling Speech Technology to 1,000+ Languages](https://mlsanity.com/paper/2305.13516/) (2023)
- [Generative Spoken Language Modeling from Raw Audio](https://mlsanity.com/paper/2102.01192/) (2021)
- [Mini-Omni: Language Models Can Hear, Talk While Thinking in Streaming](https://mlsanity.com/paper/2408.16725/) (2024)
- [Emilia: An Extensive, Multilingual, and Diverse Speech Dataset for Large-Scale Speech Generation](https://mlsanity.com/paper/2407.05361/) (2024)
- [Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations](https://mlsanity.com/paper/2405.18392/) (2024)
- [On The Landscape of Spoken Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2504.08528/) (2025)
- [WavLLM: Towards Robust and Adaptive Speech Large Language Model](https://mlsanity.com/paper/2404.00656/) (2024)
- [Textually Pretrained Speech Language Models](https://mlsanity.com/paper/2305.13009/) (2023)
- [Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context](https://mlsanity.com/paper/2309.08105/) (2023)
- [SLM: Bridge the thin gap between speech and text foundation models](https://mlsanity.com/paper/2310.00230/) (2023)
- [AudioChatLlama: Towards General-Purpose Speech Abilities for LLMs](https://mlsanity.com/paper/2311.06753/) (2023)
- [Scaling Speech-Text Pre-training with Synthetic Interleaved Data](https://mlsanity.com/paper/2411.17607/) (2024)
- [Distilling an End-to-End Voice Assistant Without Instruction Training Data](https://mlsanity.com/paper/2410.02678/) (2024)
- [Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection](https://mlsanity.com/paper/2502.06042/) (2025)
- [Scaling Properties of Speech Language Models](https://mlsanity.com/paper/2404.00685/) (2024)
- [fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit](https://mlsanity.com/paper/2109.06912/) (2021)
- [Salmon: A Suite for Acoustic Language Model Evaluation](https://mlsanity.com/paper/2409.07437/) (2024)
- [Task-Adaptive Pretrained Language Models via Clustered-Importance Sampling](https://mlsanity.com/paper/2410.03735/) (2024)
- [VoxEval: Benchmarking the Knowledge Understanding Capabilities of End-to-End Spoken Language Models](https://mlsanity.com/paper/2501.04962/) (2025)
- [TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling](https://mlsanity.com/paper/2504.07053/) (2025)
- [Scaling Analysis of Interleaved Speech-Text Language Models](https://mlsanity.com/paper/2504.02398/) (2025)

## Built on by (most cited first, 17 of 17)

- [Audio MultiChallenge: A Multi-Turn Evaluation of Spoken Dialogue Systems on Natural Human Interaction](https://mlsanity.com/paper/2512.14865/) (2025)
- [X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMs](https://mlsanity.com/paper/2603.24596/) (2026)
- [CORD: Bridging the Audio-Text Reasoning Gap via Weighted On-policy Cross-modal Distillation](https://mlsanity.com/paper/2601.16547/) (2026)
- [Closing the Modality Reasoning Gap for Speech Large Language Models](https://mlsanity.com/paper/2601.05543/) (2026)
- [The Cascade Equivalence Hypothesis: When Do Speech LLMs Behave Like ASRrightarrowLLM Pipelines?](https://mlsanity.com/paper/2602.17598/) (2026)
- [Minimizing Modality Gap from the Input Side: Your Speech LLM Can Be a Prosody-Aware Text LLM](https://mlsanity.com/paper/2605.05927/) (2026)
- [Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs](https://mlsanity.com/paper/2602.23136/) (2026)
- [Speech Codec Probing from Semantic and Phonetic Perspectives](https://mlsanity.com/paper/2603.10371/) (2026)
- [Subspace Control: Turning Constrained Model Steering into Controllable Spectral Optimization](https://mlsanity.com/paper/2604.04231/) (2026)
- [Do Factual Recall Mechanisms Carry over from Text to Speech in Multimodal Language Models?](https://mlsanity.com/paper/2605.22170/) (2026)
- [Is Text All You Need? Text as a Universal Information Bottleneck for Speech LLMs](https://mlsanity.com/paper/2606.09366/) (2026)
- [Reducing the Output-Mode Gap in Speech Language Models via Joint-Output On-Policy Distillation](https://mlsanity.com/paper/2609.15313/) (2026)
- [Transcribe, Then Reason: Two-Pass Decomposition for Multimodal Review](https://mlsanity.com/paper/2609.18958/) (2026)
- [To Trust or Not to Trust: Retrieval-Augmented Fact Checking in Speech](https://mlsanity.com/paper/2609.30227/) (2026)
- [DirectSpeech2LLM: A Simple End-to-End Framework to Mitigate Prompt Overfitting in Speech-LLMs](https://mlsanity.com/paper/2610.08085/) (2026)
- [Distilling Text LMs into Speech-to-Speech LMs by Matching Internal Text Predictions](https://mlsanity.com/paper/2610.17461/) (2026)
- [Repairing Speech Access After Text Updates with Frozen Affine Fields](https://mlsanity.com/paper/2610.20253/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) can be adapted to extend their text capabilities to speech inputs. However, these speech-adapted LLMs consistently underperform their text-based counterparts--and even cascaded pipelines--on language understanding tasks. We term this shortfall the text-speech understanding gap: the performance drop observed when a speech-adapted LLM processes spoken inputs relative to when the original text-based LLM processes the equivalent text. Recent approaches to narrowing this gap either rely on large-scale speech synthesis of text corpora, which is costly and heavily dependent on synthetic data, or on large-scale proprietary speech datasets, which are not reproducible. As a result, there remains a need for more data-efficient alternatives for closing the text-speech understanding gap. In this work, we analyze the gap as driven by two factors: (i) forgetting of text capabilities during adaptation, and (ii) cross-modal misalignment between speech and text. Based on this analysis, we introduce SALAD--Sample-efficient Alignment with Learning through Active selection and cross-modal Distillation--which combines cross-modal distillation with targeted synthetic data to improve alignment while mitigating forgetting. Applied to 3B and 7B LLMs, SALAD achieves competitive performance with a strong open-weight model across broad-domain benchmarks in knowledge, language understanding, and reasoning, while training on over an order of magnitude less speech data from public corpora.