The real time feature is very cool. But when I went through the code. Something catch my attention.
The model used in Stocktalk is nltk.sentiment.sentiment_analyzer module. In the documentation, we can see that the tool is not really state-of-the-art.
A SentimentAnalyzer is a tool to implement and facilitate Sentiment Analysis tasks using NLTK features and classifiers, especially for teaching and demonstrative purposes.
The real-time request and pyplot is very sophisticated and efficient. I would like to see if the developer can create a Stocktalk 2.0 featuring a more mature technique like wod2vec
RE: Using Python to track twitter sentiment about cryptocurrencies