Real Time AI Speech , RNN’s , Tensor Flow , Drones


Google I/O is an annual conference where Google employees and guests can compare notes and showcase the latest technological innovations . Technological innovations that we don’t need , but in years to come we will think that we need , if only because someone else has one . Attendees also drink beer , party and probably take narcotic substances , all in the interests of creativity of course .

The show stealer at this years conference highlights how far speech recognition and AI generated speech has progressed .

The system is called Google Duplex , although the conversations in the demonstrations are short , they are also realistic .

You can read more about Duplex and find examples of Duplex in action at the links below , have a listen .

There are other examples available on the web .

Go with the Flow , TI Joe

Duplex uses a recurrent neural network (RNN) , which is a type of artificial neural network (ANN) . RNN’s are inspired by the neural networks in the human brain in that they use ‘neurons’ which can signal to each other . RNN’s can also process two different sources of input , that is the real time input from a human and their own learned input , which is fed to the network via a feedback loop .

RNN’s are therefore able to learn from previous processes , it is often said that RNN’s have a ‘memory’ .

Duplex uses TensorFlow , an open source library used in many machine learning applications .

Related :


  1. A Beginner’s Guide to Recurrent Networks and LSTMs
  2. Understanding LSTM Networks
  3. Convolutional Neural Networks (CNN)
  4. Deep Neural Networks (DNN)
  5. Random Recurrent Networks Near Criticality Capture the Broadband Power Distribution of Human ECoG Dynamics (PDF)
  6. Fuzzy systems and dynamical recurrent networks for ECoG-based BCI (PDF)
  7. Understanding the neural basis of speech production using Machine Learning (PDF)
  8. Deep Learning in Bioinformatics
  9. Microarrays in the Brain: Can They Be Used for Brain–Machine Interface Control?
  10. TensorFlow

Also related :


  1. Pentagon using Google’s TensorFlow APIs to analyze drone footage
  2. Google’s AI is being used by US military drone programme
  3. AI and machine learning: A new kind of military intelligence
  4. DARPA launches PPAML artificial intelligence program
  5. China, Russia and the US Are in An Artificial Intelligence Arms Race
This entry was posted in AI, Mind Control, Science, Targeted Individuals. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s