US transportation regulator NTSB pulled its accident reports after the audio recreations were uploaded online. The National Transportation Safety Board (NTSB) has pulled its docket system offline after people used information uploaded to it to recreate the voices of pilots killed in a plane crash with AI. As CNN reports, the agency recently uploaded files filled with details about the November 4, 2025 crash involving UPS flight 2976. One of the plane's engines separated from the wing during takeoff from Louisville, Kentucky, killing three crew members and 12 people on the ground. While the NTSB uploads accident reports that the public can access, it is not allowed by federal law to release cockpit audio recordings due to the highly sensitive nature of verbal communications inside the cockpit.

Humans learn to speak before they can read or write, so why can't computers do the same? In this paper, we present a deep neural network model capable of rudimentary spoken language acquisition using untranscribed audio training data, whose only supervision comes in the form of contextually relevant visual images. We describe the collection of our data comprised of over 120,000 spoken audio captions for the Places image dataset and evaluate our model on an image search and annotation task. We also provide some visualizations which suggest that our model is learning to recognize meaningful words within the caption spectrograms.

Spectrograms are 2D representations of sound that look very different from the images found in our visual world. And natural images, when played as spectrograms, make unnatural sounds. In this paper, we show that it is possible to synthesize spectrograms that simultaneously look like natural images and sound like natural audio. We call these visual spectrograms . Our approach is simple and zero-shot, and it leverages pre-trained text-to-image and text-to-spectrogram diffusion models that operate in a shared latent space. During the reverse process, we denoise noisy latents with both the audio and image diffusion models in parallel, resulting in a sample that is likely under both models. Through quantitative evaluations and perceptual studies, we find that our method successfully generates spectrograms that align with a desired audio prompt while also taking the visual appearance of a desired image prompt.

A current gap in the acoustic space is a large-scale, fine-grained dataset of species sounds that is easy to download and easy to use.

Spectrograms are 2D representations of sound that look very different from the images found in our visual world. And natural images, when played as spectrograms, make unnatural sounds.

Neural Information Processing Systems http://nips.cc/

Neural Information Processing Systems http://nips.cc/