In this video, we're going to learn about Moonshine, a series of models optimized for real-time ASR (Automatic Speech Recognition). We learn why it's better than OpenAI's Whisper models if low latency if what we need, and then I read a TechCrunch article and we see how well it works
🟡 TechCrunch article - https://techcrunch.com/2024/11/06/the-other-election-night-winner-perplexity/
🟡 moonshine - https://github.com/usefulsensors/moonshine/tree/main
🟡 moonshine paper - https://arxiv.org/abs/2410.15608
🟡 Live captioning demo - https://github.com/usefulsensors/moonshine/blob/main/moonshine/demo/README.md#demo-live-captioning-from-microphone-input