Conversation with James Poulter (CEO of Vixen Labs) and John Kelvie
summary
Bespoken builds tooling for voice and chat that has run since 2017 and claims concrete wins: automated testing, training, monitoring for Alexa, Google, IVR and custom assistants; hands-on work improving or “unbundling” Amazon Lex for clients like Genesis, Roku and Mercedes; shows measurable ROI by offloading call‑center traffic and accelerating cloud migrations to Amazon Connect or Genesys. Common failure mode is late testing — teams often scrap Lex or Dialogflow after six months — so Bespoken pushes testing into design to catch problems early. Amazon Lex shortcomings
AI-written summary · about
chapters (13)
- 0:00Introduction and welcome
- 0:22Bespoken’s background and team
- 1:03Why voice testing matters
- 2:26Customer expectations in voice tech
- 3:29ROI and quality in contact centers
- 4:31Testing complex voice experiences
- 5:34Alexa and Google Assistant challenges
- 6:59Future of custom voice assistants
- 8:25Training and lifecycle of voice products
- 9:26Call center automation potential
- 10:49When to start testing voice systems
- 11:30Avoiding tech stack mistakes
- 11:51How to reach Bespoken
related talks
47mBuilding sophisticated conversational AI experiences
12mVOICE Summit 2019 | VED Talks | John Gillilan | July 23
50mSpeak To Me: Voice App Conversation Practices
52mThe most natural UI. Building voice-enabled experiences with Amazon Alexa
20mBuilding Smarter Cloud Support Systems with AWS Lex
02h 35mConversational Bots : Amazon Lex, IBM Watson, Microsoft Bot and Google Assistant
01h 08mChatbots - latest news
39mPotencie su centro de contacto con la inteligencia artificial