Jev in production › Message and ticket triage
SipPulse Meet Capture
Jev classifies each live Google Meet transcript line for intent, sentiment and action items in about 300 ms (company).
300 msmeasured, as published by the source
- Use
- Message and ticket triage
- Industry
- Productivity
- Form
- In-house system
- Stage
- In production
- Plugs into
- Google Meet
- Listed
- 2026-09-29
- Found via
- github
- Repository
- sippulse/meetvcon
- Stars
- 5
- Forks
- 0
- Last push
- 2026-09-29
- Language
- JavaScript
- License
- MIT
The README opens with
Internal Chrome extension for live Google Meet transcription and meeting intelligence. While the call runs, Chrome's side panel shows a live, speaker-labelled transcript, AI notes (summary, action items, topics, intents), and talk time with sentiment per speaker. When the call ends, one vCon with the transcript and meeting report is stored in SipPulse CRM and emailed to the capturing collaborator (@sippulse.com by default; the AllowedEmailDomains policy can change it for other organizations).
Badge
For the project's own README, linking back here:
Also used for message and ticket triage
- One (Inbox Run) ~580 conversations/sec
- Jev Orchestrator $0.00003
- Jevops 70-500ms
- jev-oncall 0.87
- hush (emreozyoruk) 78%
- openpoke-meets-jev (0xShin0221) 424 ms
- Surely – AI Triage (Devcore) 74%
- jev-inbox-triage 19/20