/* ---- Google Analytics Code Below */

Friday, January 12, 2018

Sound Classification

Sounds are important.  Getting them right can make the difference between correct and incorrect advice.  Those of you experimenting with advisors have now seen this.

Sound Classification with TensorFlow

Having this solution along with an IoT platform allows you to build a smart solution over a very wide area.

There are many different projects and services for human speech recognition like Pocketsphinx, Google’s Speech API, and many others. Such applications and services recognize speech to text with pretty good quality, but none of them can determine different sounds captured by the microphone. What was on record: human speech, animal sounds, or music playing?

We were faced with this task and decided to investigate and build sample projects which will be able to classify different sounds using machine learning algorithms. This article describes which tools we have chosen, what challenges we have faced, how we have trained the model for TensorFlow, and how to run our open source project. Also we can supply the recognition results to the DeviceHive, IoT platform, to use them in cloud services for 3rd party application.  .... " 

No comments: