Read one paragraph, speak freely for half a minute — our AI maps your voice fingerprint against native speakers and reveals how strong your accent really is.
Where will you land?
~2 minutes · no sign-up · instant result
You read one fixed English paragraph aloud and then speak freely for about 30 seconds. Together they capture both your careful and your natural pronunciation — around 90 seconds in total.
A speech AI model condenses your voice into a 192-dimension acoustic fingerprint and measures its distance from the center of thousands of native English speech samples.
You receive a 0-100 accent strength score, a friendly band (faint, mild, moderate or strong), your native-similarity percentage and a shareable result card.
You record two short speaking tasks. An AI speech model extracts a 192-dimension voice fingerprint from your recording and compares it with the acoustic center of native English speakers. The distance between the two becomes your 0-100 accent strength score and native-similarity percentage.
Yes — the test is completely free, takes about two minutes and requires no account or sign-up. You see your result immediately.
Your recording is used to generate your result and to improve Flalingo's accent technology. It is processed and stored only with your explicit consent, and you can request deletion at any time.
The score expresses how far your pronunciation sits from the native English reference: 0 means indistinguishable from a native speaker, 100 means a very distinct accent. An accent is not a flaw — the score simply shows how strongly it is perceived.
Your recording is processed only after you give explicit consent on the consent step. It is used to generate your accent analysis and to improve Flalingo's accent technology. You must be 18 or older to participate, and you can request deletion of your recording at any time.