Zeta - Web

Automated speech emotion classification training portal

Experience Reinforcement Learning based Domain adaptation in action with two easy steps

  1. Record two second audio segment
  2. View results and give feedback

Record two second speech audio


Record
Avoid recording sensitive information in to the audio segment.
˚
As Reinforcement Learning algorithms need significant amount of instances to re-train the model, we will only be re-training the model after collecting abundant amount of training data and Feature embedding of the recorded audio is stored until the model is re-trained.