pannous / tensorflow-speech-recognition

🎙Speech recognition using the tensorflow deep learning framework, sequence-to-sequence neural networks
Other
2.17k stars 638 forks source link

How to use this code for real-time recognition? #66

Closed JrSurapaneni closed 5 years ago

JrSurapaneni commented 5 years ago

This code is working brilliantly for saved audio files , how to make it work for real-time scenarios. Like User speaks and it recognize spoken digits.

nitishxp commented 5 years ago

@JrSurapaneni did u able to get real-time recognition?

JrSurapaneni commented 5 years ago

I trained it, prepared a model which could predict pre-recorded samples at 30% accuracy. I dont think it can be utilized as realtime model.