Introducing at16k - Open-source and offline speech recognition library

I strongly believe that there should be alternatives to Google, Amazon and Microsoft when it comes to speech recognition services. Voice is the next frontier in human-machine interfaces, and it's better if the technology and models are accessible to everyone.

With this in mind, I have released at16k, a Python library for offline speech recognition. The models are trained using a combination of openly available datasets and manually collected data from various sources (internet, field, etc.) over the past two years. The goal is to provide the community with an easy-to-use API and production quality models for building voice-based applications.

You can clone or fork us from our Github repo or try out the live demo.

Happy building!