Mixing voice & data
summary
A live demo wires a Twilio Video API WebRTC call into a Node.js backend on Bluemix that streams browser audio over WebSockets to IBM Watson Speech to Text for real-time transcription, then uses Alchemy API to extract the product and sentiment after the call, updates a 0–100 feedback score, and triggers Business Rules plus Bluemix Workflow to open a review task. Client-side code converts audio to PCM16 (painful), requires careful WebSocket handling and browser workarounds (Firefox 35, Chrome limitations).
AI-written summary · about
chapters (6)
- 0:00Start
- 3:34What is Bluemix?
- 6:52Record the local stream
- 8:24Convert the audio to PCM
- 9:20Emit a web socket message
- 9:39Initiate a websocket connection from Node to IBM Watson
related talks
19mAudio Streaming - Using WebRTC for building your own Voice AIs
43mNow Hear This! Putting Real-Time Voice, Video and Text into Rails
10mA Day in the Life: An Immersive Data Experience
01h 25mUsing Nexmo Voice WebSockets with IBM Watson
05mLive Voice Transformations For Web Conferencing Using The Browser
28mLet's Get Chatty with Conversational Interface in JavaScript
38mBuilding Smarter Applications Using Cognitive Computing
01h 04m“Hello, AI?!” – Real-Time Voice Interactions with LLMs
