044: Speech Recognition and Spoken Language Understanding (SLU)

In our second technical dive in the the anatomy of a voice assistant, Kylie hosts Shawn Wen, co-founder and CTO of Poly AI, to analyze the complexities of building voice assistants. He highlights the challenges faced in speech recognition, such as dealing with errors, latency, and user experience. Shawn also discusses the inefficiencies of converting chatbots to voice assistants and the nuances that must be managed, including managing speech recognition models, right-sizing latency, agile dialogue design, and dealing with telephony filters. He emphasizes that optimizing these systems isn't just a technical problem but a multifaceted user experience challenge.

• Follow PolyAI on LinkedIn
• Watch this and other episodes of the Deep Learning pod on YouTube