Maybot is a deep neural network that completes sentences in the style of Theresa May's parliamentary spoken contributions.
There are implementations in subdirectories for the web (maybot_web), as a skill for Amazon Alexa (maybot_alexa) and as an action for Google Home/Assistant (maybot_google). The Alexa and Google versions are called "Political Theresa" as the voice systems require a multi-word activation name.
These implementations use a common neural network model. The directory structure supports having multiple models, each as a subdirectory of the models directory. The model symlink points to the chosen model. Each model must provide a Python create_sentence() function, but can otherwise use whatever technology is desired.
The model provided, models/sjb_word_model, is written using Keras in Python. The model predicts the next word in a sentence from the preceding 12 words which are maintained in a shift register and presented to the model in parallel. The model has 4 dense layers with interspersed droput and ELU layers. word2vec is used to code words as 300-element vectors which form the input to the network. The output of the model has a probabalistic selector so that it can be adjusted by "temperature" to be more or less inventive.
The spoken contributions used for training have been obtained by scraping Hansard Online, the official verbatim report of proceedings of both the House of Commons and the House of Lords.
hansard_scrape contains the code to scrape the spoken contributions and the data files that have been collected. The key output files are all_contributions.json, other_contributions.json and theresa_may_contributions.json. These are symlinked to by the models which use them as training data.
images contains the original images and the files created from them that are used by the different web/voice platform implementations.
test.py is a basic test program to quickly check that the model is working. It completes a sentence starting "The right honourable gentleman" and prints out the result.
run.sh starts maybot_web, maybot_alexa, maybot_google and ngrok in a tabbed gnome-terminal. The ngrok settings are contained in ngrok.yml. Edit this file according to how you are going to expose the servers.