Adaptive Inner Speech-Text Alignment for LLM-based Speech Translation

• URL: http://arxiv.org/abs/2503.10211v1
• Date: Thu, 13 Mar 2025 09:54:35 GMT
• Title: Adaptive Inner Speech-Text Alignment for LLM-based Speech Translation
• Abstract summary: We propose an Adaptive Inner Speech-Text Alignment (AI-STA) method to bridge the modality gap by explicitly aligning speech and text representations at selected layers within large language models (LLMs)<n> Experimental results on speech translation tasks demonstrate that AI-STA significantly improves the translation performance of large speech-text models (LSMs), outperforming previous state-of-the-art approaches.
• Score: 20.415410280412697
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advancement of large language models (LLMs) has led to significant breakthroughs across various tasks, laying the foundation for the development of LLM-based speech translation systems. Existing methods primarily focus on aligning inputs and outputs across modalities while overlooking deeper semantic alignment within model representations. To address this limitation, we propose an Adaptive Inner Speech-Text Alignment (AI-STA) method to bridge the modality gap by explicitly aligning speech and text representations at selected layers within LLMs. To achieve this, we leverage the optimal transport (OT) theory to quantify fine-grained representation discrepancies between speech and text. Furthermore, we utilize the cross-modal retrieval technique to identify the layers that are best suited for alignment and perform joint training on these layers. Experimental results on speech translation (ST) tasks demonstrate that AI-STA significantly improves the translation performance of large speech-text models (LSMs), outperforming previous state-of-the-art approaches. Our findings highlight the importance of inner-layer speech-text alignment in LLMs and provide new insights into enhancing cross-modal learning.

Related papers

• Seg-Agent: Test-Time Multimodal Reasoning for Training-Free Language-Guided Segmentation [52.8308168727975]
  Seg-Agent is a completely training-free framework that pioneers Explicit Multimodal Chain-of-Reasoning.<n>Our approach constructs an interactive visual reasoning loop comprising three stages: generation, selection, and refinement.<n>Various-LangSeg is a novel benchmark covering explicit semantic, generic object, and reasoning-guided segmentation tasks.
  arXiv  Detail & Related papers  (2026-05-13T03:36:44Z)
• Closing the Gap Between Text and Speech Understanding in LLMs [28.538793793887223]
  Large Language Models can be adapted to extend their text capabilities to speech inputs.<n>These speech-adapted LLMs consistently underperform their text-based counterparts.<n>We introduce SALAD--Sample-efficient Alignment with Learning through Active selection and cross-modal Distillation.
  arXiv  Detail & Related papers  (2025-10-15T14:57:16Z)
• Understanding the Modality Gap: An Empirical Study on the Speech-Text Alignment Mechanism of Large Speech Language Models [12.263637152835713]
  End-to-end Large Speech Language Models (LSLMs) have demonstrated impressive conversational generation abilities.<n>We analyze both coarse- and fine-grained text and speech representations.<n>We find that representation similarity is strongly correlated with the modality gap.
  arXiv  Detail & Related papers  (2025-10-14T03:34:38Z)
• Enhancing Speech Large Language Models through Reinforced Behavior Alignment [5.647822820528311]
  This paper introduces a framework termed Reinforced Behavior Alignment (RBA) to bolster the language generation proficiency of SpeechLMs.<n>Instead of relying on supervised fine-tuning from human annotations, RBA employs a self-synthesis methodology to generate extensive, high-fidelity alignment data.<n> Experimental results demonstrate that this method effectively enhances the instruction-following capabilities of SpeechLMs.
  arXiv  Detail & Related papers  (2025-08-25T07:31:48Z)
• Optimal Transport Regularization for Speech Text Alignment in Spoken Language Models [46.76139085979338]
  OTReg is a method that formulates speech-text alignment as an optimal transport problem and derives a regularization loss to improve SLM training.<n> OTReg is lightweight, requiring no additional labels or learnable parameters, and integrates seamlessly into existing SLM training procedures.
  arXiv  Detail & Related papers  (2025-08-11T16:06:04Z)
• Adversarial Deep Metric Learning for Cross-Modal Audio-Text Alignment in Open-Vocabulary Keyword Spotting [8.401528952094413]
  For text enrollment-based open-vocabulary keyword spotting (KWS), acoustic and text embeddings are typically compared at either the phoneme or utterance level.<n>We optimize acoustic and text encoders using deep metric learning (DML), enabling direct comparison of multi-modal embeddings in a shared embedding space.<n>We propose Modality Adversarial Learning (MAL), which reduces the domain gap in heterogeneous modality representations.
  arXiv  Detail & Related papers  (2025-05-22T14:49:46Z)
• Text-Speech Language Models with Improved Cross-Modal Transfer by Aligning Abstraction Levels [12.179720395334503]
  We propose augmenting vocabulary expansion with modules that better align abstraction levels across layers.<n>Our models, textscSmolTolk, rival or surpass state-of-the-art TSLMs trained with orders of magnitude more compute.
  arXiv  Detail & Related papers  (2025-03-08T13:28:50Z)
• SparQLe: Speech Queries to Text Translation Through LLMs [0.8901073744693314]
  This study introduces a novel approach that leverages self-supervised speech representations in combination with instruction-tuned LLMs for speech-to-text translation.<n>Our experiments demonstrate that this method effectively preserves the semantic content of the input speech and serves as an effective bridge between self-supervised speech models and instruction-tuned LLMs.
  arXiv  Detail & Related papers  (2025-02-13T12:57:15Z)
• AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Understanding [63.09928907734156]
  AlignVLM is a vision-text alignment method that maps visual features to a weighted average of text embeddings.<n>Our experiments show that AlignVLM achieves state-of-the-art performance compared to prior alignment methods.
  arXiv  Detail & Related papers  (2025-02-03T13:34:51Z)
• SEAL: Speech Embedding Alignment Learning for Speech Large Language Model with Retrieval-Augmented Generation [10.828717295018123]
  We propose a unified embedding framework that eliminates the need for intermediate text representations.<n>Our model reduces pipeline latency by 50% while achieving higher retrieval accuracy compared to traditional two-stage methods.
  arXiv  Detail & Related papers  (2025-01-26T15:04:02Z)
• Align-SLM: Textless Spoken Language Models with Reinforcement Learning from AI Feedback [50.84142264245052]
  This work introduces the Align-SLM framework to enhance the semantic understanding of textless Spoken Language Models (SLMs) Our approach generates multiple speech continuations from a given prompt and uses semantic metrics to create preference data for Direct Preference Optimization (DPO) We evaluate the framework using ZeroSpeech 2021 benchmarks for lexical and syntactic modeling, the spoken version of the StoryCloze dataset for semantic coherence, and other speech generation metrics, including the GPT4-o score and human evaluation.
  arXiv  Detail & Related papers  (2024-11-04T06:07:53Z)
• Mind the Gap: A Generalized Approach for Cross-Modal Embedding Alignment [0.0]
  Retrieval-Augmented Generation (RAG) systems retrieve context across different text modalities due to semantic gaps. We introduce a generalized projection-based method, inspired by adapter modules in transfer learning, that efficiently bridges these gaps. Our approach emphasizes speed, accuracy, and data efficiency, requiring minimal resources for training and inference.
  arXiv  Detail & Related papers  (2024-10-30T20:28:10Z)
• Self-Powered LLM Modality Expansion for Large Speech-Text Models [62.27700381806554]
  Large language models (LLMs) exhibit remarkable performance across diverse tasks. This study aims to refine the use of speech datasets for LSM training by addressing the limitations of vanilla instruction tuning. We introduce a self-powered LSM that leverages augmented automatic speech recognition data generated by the model itself for more effective instruction tuning.
  arXiv  Detail & Related papers  (2024-10-04T04:34:24Z)
• Towards ASR Robust Spoken Language Understanding Through In-Context Learning With Word Confusion Networks [68.79880423713597]
  We introduce a method that utilizes the ASR system's lattice output instead of relying solely on the top hypothesis. Our in-context learning experiments, covering spoken question answering and intent classification, underline the LLM's resilience to noisy speech transcripts.
  arXiv  Detail & Related papers  (2024-01-05T17:58:10Z)
• Generative Context-aware Fine-tuning of Self-supervised Speech Models [54.389711404209415]
  We study the use of generative large language models (LLM) generated context information. We propose an approach to distill the generated information during fine-tuning of self-supervised speech models. We evaluate the proposed approach using the SLUE and Libri-light benchmarks for several downstream tasks: automatic speech recognition, named entity recognition, and sentiment analysis.
  arXiv  Detail & Related papers  (2023-12-15T15:46:02Z)
• Large Language Models can Contrastively Refine their Generation for Better Sentence Representation Learning [57.74233319453229]
  Large language models (LLMs) have emerged as a groundbreaking technology and their unparalleled text generation capabilities have sparked interest in their application to the fundamental sentence representation learning task. We propose MultiCSR, a multi-level contrastive sentence representation learning framework that decomposes the process of prompting LLMs to generate a corpus. Our experiments reveal that MultiCSR enables a less advanced LLM to surpass the performance of ChatGPT, while applying it to ChatGPT achieves better state-of-the-art results.
  arXiv  Detail & Related papers  (2023-10-17T03:21:43Z)
• ROSITA: Enhancing Vision-and-Language Semantic Alignments via Cross- and Intra-modal Knowledge Integration [48.01536973731182]
  We introduce a new vision-and-language pretraining method called ROSITA. It integrates the cross- and intra-modal knowledge in a unified scene graph to enhance the semantic alignments. ROSITA significantly outperforms existing state-of-the-art methods on three typical vision-and-language tasks over six benchmark datasets.
  arXiv  Detail & Related papers  (2021-08-16T13:16:58Z)

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.