OpenAI has open sourced its multilingual speech recognition system Whisper , and the source code is hosted on GitHub under the MIT license. Speech recognition systems are a core part of the tech giant’s software and services, and Whispe translates speech into text in multiple languages and translates it into English. It was trained using 680,000 hours of multilingual speech data, collected from the network, to improve the recognition of unique spoken words, background noise, and technical terms.
This article is reprinted from: https://www.solidot.org/story?sid=72878
This site is for inclusion only, and the copyright belongs to the original author.