Vision-to-Voice: Enhanced CNN-LSTM-Based Image Captioning with Assistive Text-to-Speech

Image captioning is a multidisciplinary task that combines the capabilities of computer vision and natural language processing, enabling automatic generation of descriptive text for images. This paper presents an approach that leverages Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks for enhanced image captioning, trained on the rich Flickr8k dataset, which provides diverse real-world images with multiple humanannotated captions. Our work differentiates itself by optimizing feature extraction and sequence generation to improve contextual accuracy and fluency. CNNs extract essential visual features, which are then processed by an LSTM-based decoder to generate coherent and meaningful captions while retaining contextual information. Additionally, we introduce an assistive text-to-voice feature that reads out the generated captions, making the system more accessible for the visually impaired. Experimental results demonstrate improved caption quality compared to existing approaches. This framework has broad applications, from assistive technologies to multimedia content enrichment, further advancing semantic understanding and human-computer interactions.

• Publication year

  2025

• Venue

  2025 IEEE 2nd International Conference on Green Industrial Electronics and Sustainable Technologies (GIEST)

• Publication date

  2025-10-11

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/GIEST66547.2025.11387720
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Image caption generator using CNN & LSTM

  2023cited by this paper
• Domain-Specific Image Caption Generator with Semantic Ontology

  2020cited by this paper
• Towards Personalized Image Captioning via Multimodal Memory Networks

  2019cited by this paper
• Automatic Image Captioning Using Convolution Neural Networks and LSTM

  2019cited by this paper
• Image Captioning: Transforming Objects into Words

  2019cited by this paper
• A survey on automatic image caption generation

  2018cited by this paper
• Regularizing RNNs for Caption Generation by Reconstructing the Past with the Present

  2018cited by this paper
• SemStyle: Learning to Generate Stylised Image Captions Using Unaligned Text

  2018cited by this paper
• Convolutional Image Captioning

  2017cited by this paper
• Deep learning in big data Analytics: A comparative study

  2017cited by this paper
• Boosting Image Captioning with Attributes

  2016cited by this paper
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention

  2015cited by this paper
• Show and tell: A neural image caption generator

  2014cited by this paper
• Neural Machine Translation by Jointly Learning to Align and Translate

  2014cited by this paper
• Novel neural modulators.

  2003cited by this paper

• No citing papers are available for this paper.