METHOD · 05 · PRONUNCIATION
Hear it before you commit.
Brand names live as sound — in pitches, in voice search, in customer-service calls, in podcast intros. Etymolt is the only brand pronunciation test grounded in actual acoustic reality: TTS to Whisper STT round-trip across twelve accents, scored.
Type a name · hear it in five accents
THREE THINGS
What the acoustic axis resolves.
FOR FOUNDERS
The finalist decider
Stuck between three names? Hear each in your actual elevator pitch. The dead one becomes obvious in 30 seconds.
FOR VERDICTS
Pronunciation Resilience
We render your name in 5 voices, run it back through Whisper STT, and measure character-error-rate. A new scoring axis no competitor has.
FOR FORGE ENGAGEMENTS
Sonic logo + audio dossier
Every Forge engagement — the post-verdict execution surface — includes a 3-second sonic logo and a 90-second audio narration of your verdict. Marketing-ready from day one.
METHOD
The Pronunciation Resilience methodology.
Most naming tools stop at text. We measure how a name behaves when it becomes sound.
Render the brand name in five distinct voices (US English, British, US Narrator, US Warm, US Confident). Submit each rendering to OpenAI Whisper STT. Compute character-error-rate between the input and the transcription. Aggregate with per-persona weights. Map to a 0–100 Pronunciation Resilience score.
Names that consistently misclass under Whisper will similarly trip up voice search, voicemail transcription, IVR menus, and live captions. The score is an indirect but consistent measurement of real-world voice-channel risk.
Full methodology with calibration eval set, references, and BibTeX export at /methodology#pronunciation.
The voice preview is one signal.
The full Etymolt verdict gives you four more — sound symbolism, USPTO clearance, cultural screen across 20 markets, phonetic neighbors. One verdict. Permalink. Free.
Naming, attested.