Get the latest tech news

Auto-AVSR: Audio-Visual Speech Recognition with Automatic Labels


Auto-AVSR: Lip-Reading Sentences Project. Contribute to mpc001/auto_avsr development by creating an account on GitHub.

It is designed for end-to-end training, aiming to deliver state-of-the-art models and enable reproducibility on audio-visual speech benchmarks. Required arguments exp-dir: Directory to save checkpoints and logs to, default:./exp. The pre-trained models provided in this repository may have their own licenses or terms and conditions derived from the dataset used for training.

Get the Android app

Or read this on Hacker News

Read more on:

Photo of automatic labels

automatic labels

Photo of avsr

avsr