Paste a copied X/Twitter thread, a hashtag feed, or a chat export below. Signal Desk splits it into posts, guesses roles, scores sentiment and vibe, and builds a dashboard โ then a studio for turning it into promo, article, flashcard, quiz, and site content.
FORMAT
EXAMPLE FEEDS
OR UPLOAD SCREENSHOTS / FILES
Screenshots get read by Claude โ text, handles, and the general vibe/energy of any memes or images get pulled out and dropped into the box below, ready to parse. Works one screenshot at a time, so give it a second per image.
No feed parsed yet. Head to Intake and paste something in.
Voices
Timeline โ click text to fix a misparse, tap a reaction to vibe
Post-by-post sentiment
BullishNeutralBearishhover a cell for the post
Vibe by voice
Trending words & phrases
Parse a feed first โ the studio writes from what's in your dashboard.
TONELANGUAGE
Live research โ optional, pulls current web context into your generations
No research pulled yet โ every content type below will just use the thread itself until you run one.
Saved
Generated copy is a creative starting point, not financial advice โ keep any claims about a token, project, or trade honest before you post it, including anything pulled from live research. Reaction counts on the Timeline are shared with anyone who opens this artifact.
Parse a feed first โ Theater performs it back to you.
Every voice gets a generated avatar (abstract, not a real photo of anyone) and a distinct synthesized voice. Hit play and watch the thread perform itself, one post at a time โ think table-read, not full animation.
โ
Press play to start the read-through.
SPEED0 / 0
Paste your own ElevenLabs API key to swap the free browser voice for a real neural voice. The key stays in this browser tab only โ it is never saved to storage, never sent anywhere but ElevenLabs' own API, and disappears if you reload. Get one at elevenlabs.io if you don't have one; leave this blank to keep using the default free voices.
CLONE A VOICE
Upload a clean ~10โ30 second recording of a voice you own or have explicit permission to clone. Real-time, actual voice cloning โ not a simulation.
Only clone voices you own or have explicit permission to use. Don't clone a real public figure to put words in their mouth โ that's not what this is for.
If you have Voicebox (which can run Chatterbox as one of its engines) running on this computer, this tries to talk to it directly at its local address โ completely free, nothing leaves your machine. Heads up: browsers often block a hosted page like this one from reaching localhost for security reasons, so this may simply refuse to connect โ that's a platform boundary, not something wrong with your setup.
USE PROFILE(applies to every speaker while local engine is selected)
Voices are synthesized (browser TTS, your own ElevenLabs key, a voice you cloned yourself, or your own local Voicebox/Chatterbox engine) โ not real recordings of anyone unless you made them, and not affiliated with the people quoted in a thread. Have fun with it, but don't pass generated audio off as something someone actually said out loud.