Open-source, local and private voice assistant that clears your inbox and books your meetings
The Hook
“(No spoken dialogue — ambient subway train sound effects)”
Product Reveal
“Hey Jo, nice to hear you. I cleared 12 promotional emails that came in for you overnight. TJ wants a meeting about the new project. / Tell him totally. Look for any available spots in my calendar for Thursday.”
Feature Teaser
“Meeting booked to your calendar. Maya needs sign-off on the board deck. / Okay, draft a response to her email and tell her that Nate approved it last night.”
Feature Teaser
“Drafted in your email. Nate's asking if the deck's final on Slack. / Yes. Tell him it's final and the team did a great job on it.”
Call to Action
“Sent Nate your reply on Slack. / Thank you, Jo. / Codename Jo. Do work with your voice.”
Synthesizable in Remotion · key components of the voice-assistant-lifestyle-walkthrough template.
Translucent rounded rectangular HUD pill with audio sphere glyph that expands dynamically as transcription updates.
Use Remotion spring on opacity and translateY with backdropFilter blur and linear layout padding.
Metallic 3D orb paired with san-serif brand name and subline fading upward in sync with audio cue.
Interpolate opacity from 0 to 1 and translateY from 12px to 0px using spring animation with a gentle damping.
Purpose. Establish commute context and introduce hands-free AI interaction.
Execution. Transit cuts leading to woman seated in train with conversational overlay.
Purpose. Demonstrate seamless multi-app triage across Calendar, Email, and Slack while moving.
Execution. Continuous pedestrian tracking through station and street with real-time UI bubbles.
Purpose. Confirm task completion and reinforce core value proposition.
Execution. Close-up on phone interface resolving to outdoor wide shot with brand title lockup.
Objective
Launch of a voice-activated developer assistant for triaging GitHub issues and CI/CD pipelines while on the go.How the recipe adapts
Replace subway commute footage with a modern campus walk, updating speech bubbles to reflect GitHub issue summaries, pull request reviews, and Slack deployments.Scene durations (5 scenes, 36s)
Key Remotion components
Introducing Griffin, the first AI to pass the video Turing test
The most emotive and controllable TTS model
We just released QVAC TranscribePsy, a speech-to-text model small enough to run inside a phone app. It covers 25 European languages on a laptop and 8 on a phone, and on an iPhone 17 Pro Max's Ne