AudioViewer: Learning to Visualize Sounds
This research addresses the significant challenge of enabling sound perception for deaf and hard of hearing (DHH) people through visual means, offering an incremental step towards sensory substitution.
The paper tackles the problem of visualizing audio content for deaf and hard of hearing individuals by developing a model that translates generic environment sounds and human speech into video. It achieves this without paired labels by learning a mapping from unpaired audio-video examples using high-level constraints, and for speech, it disentangles content from style. The generated videos maintain important audio features, allowing humans to match and distinguish between sounds and words.
A long-standing goal in the field of sensory substitution is to enable sound perception for deaf and hard of hearing (DHH) people by visualizing audio content. Different from existing models that translate to hand sign language, between speech and text, or text and images, we target immediate and low-level audio to video translation that applies to generic environment sounds as well as human speech. Since such a substitution is artificial, without labels for supervised learning, our core contribution is to build a mapping from audio to video that learns from unpaired examples via high-level constraints. For speech, we additionally disentangle content from style, such as gender and dialect. Qualitative and quantitative results, including a human study, demonstrate that our unpaired translation approach maintains important audio features in the generated video and that videos of faces and numbers are well suited for visualizing high-dimensional audio features that can be parsed by humans to match and distinguish between sounds and words. Code and models are available at https://chunjinsong.github.io/audioviewer