Summary of Tell What You Hear From What You See — Video to Audio Generation Through Text, by Xiulong Liu and Kun Su and Eli Shlizerman
Tell What You Hear From What You See – Video to Audio Generation Through Text
by Xiulong Liu, Kun Su, Eli Shlizerman
First submitted to arxiv on: 8 Nov 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD); Audio and Speech Processing (eess.AS)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed VATT framework is a multi-modal generative model that takes a video and optional text prompt as input to generate audio and textual description. It consists of two key modules: the VATT Converter, which maps video features to an LLM vector space, and the VATT Audio transformer, which generates audio tokens from visual frames and text prompts. The framework achieves competitive performance in objective metrics and is preferred by subjective studies for generating high-quality audio. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The VATT framework generates audio based on a video and optional text prompt. This means you can control what’s said in the audio with your own words! It also suggests what audio to generate for the video, which could be useful for things like describing videos or creating audio descriptions for people who are visually impaired. |
Keywords
» Artificial intelligence » Generative model » Multi modal » Prompt » Transformer » Vector space
