Describe it from memory — I'll figure out what it is.
You know the thing — you can almost see it, taste it, hear it — but you can't name it. Describe whatever fragment you remember and AI cross-references sensory details, context, and vibes to identify it. Works for food, music, films, products, scents, colors, places, fabrics, and anything else. Refine mode narrows results based on your feedback.
TipOfTongue takes fragmentary, vibes-based descriptions and identifies what you're thinking of. Pick a category (food, music, film, product, scent, color, place, fabric, or other), describe what you remember however you can, and AI cross-references sensory details, time/place context, and elimination clues to find ranked matches. Each match includes why it fits, a memory trigger to confirm it, how to verify, and where to find it. If the first round is close but not right, refine mode uses your yes/no/close feedback to dramatically narrow the search.
Scenario: You ate something incredible at a restaurant two years ago — it was a cold noodle dish, slightly sweet and nutty, with crispy things on top. Asian but you're not sure which cuisine.
What you do: Select 'Food', describe the memory. Add context: 'ate at a trendy restaurant in Brooklyn, summer 2023'.
Result: Top match: Sichuan cold sesame noodles (liang mian) — confidence 75%. The nutty flavor is tahini/sesame paste, crispy things likely fried shallots or peanuts. Memory trigger: 'If the sauce was thick and clung to the noodles rather than pooling at the bottom, this is it.' How to find: Dan Dan Noodles at Xi'an Famous Foods, or recipe with 3 tbsp tahini + soy + chili oil + rice vinegar.
Describe everything you remember — the vibe, roughly what it means, what it's definitely not, where you encountered it — and Tip of Tongue cross-references the fragments to give you ranked candidates with confidence levels. Fragmentary, sensory, half-wrong descriptions are exactly what it's built for.
Yes — 'that movie where…' and 'that song that goes…' are its most common jobs. Give it the era, the mood, a scene you remember, an actor's face you can picture, or misremembered lyrics; it returns matches with the specific detail that will make you go 'YES, that's it', plus how to verify each one.
Words and phrases, movies and shows, songs, foods and dishes, products, scents, colors, places, fabrics — anything you can half-describe. Each category cross-references differently: food by flavor and texture, music by era and vocal style, products by packaging and where you bought it.
Mark what was close and what was wrong, add any new fragment that surfaced, and refine — 'close' answers are gold because they tell it which direction to pivot. Two or three rounds of refinement usually corner the answer for genuinely obscure memories.
The tip-of-the-tongue state is a retrieval failure, not a memory loss — your brain has the word but the phonological pathway to say it is temporarily blocked, which is why you can often recall the first letter or syllable count. Describing around the word (exactly what this tool asks you to do) activates neighboring pathways and is one of the reliable ways to spring it loose.