OpenAI’s AI model automatically translates speech from Spanish to English
OpenAI released on Wednesday a new AI model that is open-source called Whisper. It recognizes and interprets audio to a level similar to human recognition abilities. It can transcribe conversations, podcasts and interviews.
OpenAI trained Whisper using 680,000 hours audio data and matching transcriptions in 98 different languages that were collected from the internet. OpenAI claims that this approach to open collection has improved \”robustness to accents and background noise\” and also translated the spoken language into English.
Source:
https://arstechnica.com/information-technology/2022/09/new-ai-model-from-openai-automatically-recognizes-speech-and-translates-to-english/
Leave a Reply