# SVLA: A Unified Speech-Vision-Language Assistant with Multimodal Reasoning and Speech Generation

Source: [SVLA: A Unified Speech-Vision-Language Assistant with Multimodal Reasoning and Speech Generation](<https://www.sota2.com/research/paper/68fa79f11e209131cea80283>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:08:30.960Z

Authors: Ngoc Dung Huynh, Mohamed Reda Bouadjenek, Imran Razzak, Hakim Hacid, Sunil Aryal

Published: 2025-03-31T14:46:34.000Z

## Abstract

Large vision and language models show strong performance in tasks like image captioning, visual question answering, and retrieval. However, challenges remain in integrating speech, text, and vision into a unified model, especially for spoken tasks. Speech generation methods vary (some produce speech directly), others through text (but their impact on quality is unclear). Evaluation often relies on automatic speech recognition, which may introduce bias. We propose SVLA, a unified speech vision language model based on a transformer architecture that handles multimodal inputs and outputs. We train it on 38.2 million speech text image examples, including 64.1 hours of synthetic speech. We also introduce Speech VQA Accuracy, a new metric for evaluating spoken responses. SVLA improves multimodal understanding and generation by better combining speech, vision, and language.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2503.24164>)

[PDF](<https://arxiv.org/pdf/2503.24164.pdf>)
