Pythian develops prototype for toxic sentiment detection for online game developer

Customer Success

Pythian develops prototype for toxic sentiment detection for online game developer 

One of the largest developers of multiplayer online games needed to detect toxicity in gameplay audio conversations. Pythian leveraged Google Speech to Text API and Tensorflow to build a solution. 

Overview

Toxic conversations between players is a widespread issue in the online gaming world. One of the largest developers of multiplayer online games turned to Google and Pythian for assistance with a solution for detecting toxicity in gameplay conversations by analyzing the audio stream. 

The company has existing toxicity detection models that work well with instant messaging chats. However, they wanted to expand those chat detection models into online speech conversations. Pythian developed a prototype solution.  

What we did

  • Leveraged a combination of Google Speech to Text API and Tensorflow to build a gameplay transcription and toxicity detection prototype 
  • Optimized recognition of gaming slang 
  • Leveraged transfer learning on the publicly available toxicity content and gaming slang

Technologies used

  • Google Cloud Platform 
  • Google Text to Speech API 
  • Google Cloud Storage 
  • Google Kubernetes Engine  
  • Google Cloud Memorystore 
  • Google StackDriver 
  • TensorFlow

Key Outcomes

The company can now expand its natural language toxicity detection capabilities into audio gameplay conversations. 


Built a solution to evaluate the transcribed results against in-house data to assess the sentiment 


Increased accuracy of toxicity sentiments by aggregating several transcriptions of lower accuracy using different recognition engines 

Explore our Data Science, Artificial Intelligence, and Machine Learning Services   

No matter your business, no matter the challenge: Pythian’s solutions drive results. 

More customer success stories