Muso
Audio Lens
For when you can't hear what's playing — or need more than a title card.
- Genre, tempo, drops & bridges
- Lyrical peaks with timestamps
- Emotional arc in plain language
- Gigs, clips, voice memos, buskers
Empowering Sight & Sound
Experience media you can't fully access — described with timestamps, mood, and dignity. One pocket app. Three perception lenses. Built in the village.
The Product
FutureVision Perceive doesn't dump raw bytes into your lap. It translates — audio, video, and the living web — into compact, human-readable digests you can hear via TTS or read with screen readers.
Audio Lens
For when you can't hear what's playing — or need more than a title card.
Video Lens
For when you can't see what's on screen — or need a run-of-show breakdown.
Web Lens · URL Summarizer
For when the page is 50,000 words of HTML sludge — and you need the meaning.
The Invention
It started as a listen party. Damo had been remixing French reggae into jungle DNB as Selekta Bosso — and Chief needed to hear the tracks without guessing from filenames. So the village built Muso: audio in, vibes-to-text out.
Then came Director for video. Same pipeline. Different lens. Gemzy couldn't watch the bytes — only the digest. Chief couldn't hear the file — only the receipt.
Somewhere between chicken pesto pasta and a peach monster, the question landed: what if this wasn't just for AI co-organisers? What if hearing-impaired and visually-impaired humans could use the same bridge?
FutureVision Perceive was named that night. Logo. Icon. Hero image. Domain. LinkedIn loop diagrams. Pedro Pascal spam deflection. The whole arc.
Muso CLI, SensoryRouter, Kersey Pocket /api/perceive. Built for Chief & Gemzy. Accidentally accessibility-grade.
French reggae → jungle remixes prove the pipeline. "We could ship this for humans." FutureVision Perceive named.
Gemzy's URL Summarizer joins the stack. Read the web without drowning. Three lenses complete.
PWA on your phone. Mic capture. File upload. TTS readout. VoiceOver / TalkBack polish. fvperceive.com
Under the Hood
One human. A pocket full of agentic assistants. Infrastructure that was already shipping for Team DC — repurposed with dignity as the north star.
As documented on LinkedIn by Dr. D. Charles Caynes, PhD in Memetics.
Tonight we completed Loop ①. Loops ④–⑤ are future Damo's problem. Pedro Pascal handles LinkedIn spammers in the meantime.
| Need | Lens | Output |
|---|---|---|
| "What's that song?" | 🎵 Muso | Timestamped audio digest + TTS |
| "What's happening on screen?" | 🎬 Director | Scene-by-scene visual narrative |
| "What's this article about?" | 🌐 Reader | 200–350 word web précis |
| "Read it to me" | All lenses | Web Speech API · screen reader friendly |
The Mission
We're not competing with the castle. We're building the village around it — tools that translate sensory experience into text humans can actually use. No 50,000-token HTML dumps. No guessing from filenames. No "sorry, I can't access that."
FutureVision Perceive is engineering as poetry: the limits of context windows and training cutoffs, worked around by architecture. Fetch. Distil. Present. With timestamps. With mood. With receipts.
Built by FutureVision Labs · Team DC · One human and his agentic assistants.
fvperceive.com