Building a cloud AutoML platform for creating voice robots
summary
Describes building a cloud AutoML platform for creating voice robots using NLU technologies. Covers using out-of-the-box systems (DialogFlow), open-source frameworks (Rasa) and fine-tuning pretrained models like BERT for text classification, and explains a platform approach to automate and scale those ML workflows for conversational agents.
AI-written summary · about
chapters (13)
- 0:00AutoML platform introduction
- 1:10NLP technology evolution
- 10:19AutoML market challenges
- 12:09Voice bot types
- 15:40Intent classifier architecture
- 17:25First prototype iteration
- 22:44Model scaling problem
- 23:48Scaling solutions
- 26:57Hybrid transformer inference
- 32:17Hyperparameter optimization
- 35:32Multilingual support
- 37:18Project summary
- 39:52Q&A
related talks
47mBuilding smarter solutions without machine learning expertise
01h 29mGoing Beyond “Sorry, I Didn’t Get That”: Building AI Assistants that Scale...
30mConversational AI: Building clever chatbots
47mBuilding smarter apps with Machine Learning, from magic to reality
05mMachine Learning for Developers
47mTesting Conversational AI
01h 38mConversational AI with Rasa Core & NLU
54mBuilding smarter apps with Machine Learning, from magic to reality