Q n A with Sohan Maheshwar
summary
Sohan Maheshwar discusses speech recognition techniques such as hypothesis re-rankers, recurrent neural networks (RNNs) and bi-directional LSTMs, and explores applications of speech-to-text technology. He covers the transition from Q&A systems to true dialog systems, Alexa's meaning representation language, open APIs/SDKs for building skills, and his experience on the ODSC India program committee.
AI-written summary · about
related talks
26mInterview
46mIntroduction to Speech Recognition Systems Through a Tester’s Eyes
04mMike Tamir Interview
42mdata.bythebay.io: Shubho Sengupta,
27mApplying ML at scale for New User Experiences
56mSpeech Recognition: How to Build Speech-to-Text Yourself
50mCognitive Services Extravaganza!
43mHarnessing DeepSeek-R1 and OpenSearch for RAG-Powered Intelligent Info...