Researchers at Facebook AI recently introduced and open-sourced a new framework for self-supervised learning of representations from raw audio data known as wav2vec 2.0. The company claims that this framework can enable automatic speech recognition models with just 10 minutes of transcribed speech data.

Neural network models have gained much traction over the last few years due to its applications across various sectors. The models work with the help of vast quantities of labelled training data. However, most of the time, it is challenging to gather labelled data than unlabelled data.

The current speech recognition systems require thousands of hours of transcribed speech to reach acceptable performance. There are around 7,000 languages in the world and many more dialects. It can be said that the availability of the transcribed speech for a vast majority of languages is still negative.

To mitigate such issues, researchers open-sourced the wave2vec framework. The framework has the capability to make efficient development in Automatic Speech Recognition (ASR) for the low-resource languages.).

#developers corner #facebook ai #facebook ai research #speech recognition algorithm