Korean pronunciation checker. Record yourself and AI rates aspirated and tensed consonants, vowels, and rhythm.
Select the AI model for audio analysis. Different models may have different capabilities.
Record audio directly from your microphone
To check your Korean pronunciation, record yourself speaking and upload it: the AI scores you out of 10 and targets the contrasts that define learner Korean, starting with the three-way consonant distinction English completely lacks: plain, aspirated, and tense consonants (ㄱ versus ㅋ versus ㄲ and their siblings), where tense consonants require a throat tightness no English sound prepares you for. It then evaluates the vowel pairs learners merge (ㅓ versus ㅗ, ㅡ versus ㅜ, ㅐ versus ㅔ), final consonant pronunciation (받침, where stops go unreleased), and liaison and assimilation between syllables, the rules that make written and spoken Korean diverge so sharply. Intonation gets checked too, including the rise at the end of yes-no questions. You finish with three drills for the week aimed at your weakest contrasts.
Record yourself speaking Korean (read sentences plus a stretch of free speech) and upload it. The AI scores overall pronunciation, then works through the Korean-specific systems. Consonant analysis tests the three-way contrast series by series: plain ㄱ, aspirated ㅋ with its puff of air, and tense ㄲ with constricted throat and no aspiration, checking whether your three are genuinely distinct or collapsing into two. Vowel analysis targets the pairs that cost learners meaning: ㅓ against ㅗ, ㅡ against ㅜ, and ㅐ against ㅔ. Batchim analysis checks final consonants, including unreleased stops and which sounds neutralize in final position. Connected-speech analysis evaluates liaison (final consonants carrying into a following vowel) and assimilation rules that change sounds across syllable boundaries. Intonation analysis confirms yes-no questions rise at the end. The three weekly drills focus on your specific collapsing contrasts.
Record yourself speaking Korean and the checker scores you out of 10 with feedback on the contrasts that define learner Korean: plain versus aspirated versus tense consonants, the vowel pairs that tend to merge, final-consonant batchim, and connected-speech sound changes. Three weekly drills target your specific weak contrasts.
Yes, each consonant series is tested for three-way distinctness, and the most common finding is that a learner's three have collapsed into two: the aspirated version comes through but plain and tense merge. The feedback names which pairs are merging, because producing the tense series takes a throat setting English never asks for.
Sound-change rules: a final consonant slides onto a following vowel, and neighboring sounds reshape each other across syllable boundaries. The connected-speech analysis checks whether you apply these rules or pronounce each syllable exactly as written, which is one of the clearest markers of classroom Korean.
Final consonants get their own section: whether your stops are correctly unreleased (no puff of air at the end of 밥), and whether sounds neutralize in final position the way Korean requires. Released, English-style word endings are one of the fastest tells of a beginner accent.
The pairs the analysis prioritizes: ㅓ versus ㅗ, ㅡ versus ㅜ, and ㅐ versus ㅔ. The last pair has largely merged for younger native speakers, so it costs little, but ㅓ and ㅗ swap meaning in everyday words and stay high priority. The feedback shows which pairs you separate cleanly and which collapse.
The three-way consonant contrast is subtle even for trained ears, so treat borderline calls there with some skepticism; clear audio and deliberate minimal pairs (가, 카, 까 in sequence) sharpen it considerably. Vowels, batchim, and connected speech read more reliably. It evaluates sound only, not grammar, spelling, or vocabulary.
Vocal range tester. Sing your lowest and highest notes and AI tells you your range and voice type.
Baby cry translator. Record your baby and AI estimates whether the cry signals hunger, tiredness, discomfort,…
Celebrity voice matcher. Upload a voice clip and AI tells you which famous voices yours resembles.
Voice attractiveness analyzer. Upload your voice and AI rates how appealing it sounds based on research-backe…
Voice charisma analyzer. Upload a clip and AI scores warmth, magnetism, and crowd appeal.
Voice authority analyzer. Upload a clip and AI rates command presence, gravitas, and credibility.