High-precision sound analysis is our passion
DeepSound is a deep tech company specializing in high-precision sound analysis and classification using the unique neural network technology DeepSound AI.
We are engaged in the development of cutting-edge technologies and the integration of our own solutions into production and service processes, medicine, and veterinary.
* Mean average precision score
** The number of 8-seconds processed audio recordings per second
We've used the AudioSet dataset with 5.8 thousand hours of audio and 527 different classes of annotated sounds for algorithm comparison. The test was made on NVIDIA Tesla V100 16Gb graphic card.