Greg Whiteside, CEO, Human First, joins us on our podcast to tell us about how you can use different types of data to build better bots.
Presented by Deepgram
Deepgram is a Speech Company whose goal is to have every voice heard and understood. We have revolutionized speech-to-text (STT) with an End-to-End Deep Learning platform. This AI architectural advantage means you don’t have to compromise on speed, accuracy, scalability, or cost to build the next big idea in voice. Our easy-to-use SDKs and APIs allow developers to quickly test and embed our STT solution into their voice products. For more information, visit:
Using data to build better bots
Greg Whiteside, CEO, Human First joins us to share how you can use data to build better bots, why we’re focusing too much on data labelling and why your help centres and knowledge bases are underrated goldmines.
00:00 Introduction and presenting Deepgram
00:00 Introduction and presenting Deepgram
02:50 Welcome Greg Whiteside
09:35 Prior startups before HumanFirst
16:20 Fitting HumanFirst into conversational AI workflows
18:30 Why use HumanFirst?
20:55 Finding conflicts in NLU data
23:00 Utilising unstructured/conversational data
25:43 Where to start with conversational data
29:19 The sweet spot for HumanFirst
31:58 Opportunities people are missing out on
39:03 Finding value in different types of conversations
44:30 Thoughts on large language models
50:50 FAQ experiences
54:50 3 quickfire questions
58:09 Outro
Visit humanfirst.ai to learn more