Demo Session
summary
A live CodePen builds a working voice-enabled product filter in minutes using Speechly’s JS libs, dashboard and quickstart. The page adds a push-to-talk button plus a big transcript and error panel, defines two entities (color, brand) in Speechly Annotation Language, deploys a model (training ~1–2 minutes), and wires real-time callbacks to update dropdowns. The adapted ASR+NLU generalizes to unseen values, accepts complex utterances like “white sneakers by Adidas,” and the demo successfully runs end-to-end.
AI-written summary · about
chapters (9)
- 0:00Start
- 2:11Reactive Voice Uis
- 3:24Examples of Speechly in Action
- 6:18Ui Components
- 8:18Get a Speechly App Id
- 9:09Define the Indents and Entities
- 9:46Configuration of Speechly
- 10:21Variables
- 16:33Training Examples
related talks
52mGoing Full Duplex - Building Apps for Speech
10mA Day in the Life: An Immersive Data Experience
17mBuilding Effective Voice Agents
40mBuild advanced Voice Assistants and ChatBots
50mSpeak To Me: Voice App Conversation Practices
19m"Bringing Voice First Experiences to the Web"
01h 26mBuilding voice agents with OpenAI
01h 10mSolutions prototyping using Azure AI Studio