Localeit-IT
Scored dimensionsPronunciation · Completeness
LevelsSentence · word · phoneme
InputAudio ≤ 30 s + reference text

Try it live

Read the sentence below aloud and stop. The result is the real Italian engine's answer: a score for the sentence and every word, and the sounds it expected vs heard. Free, no account, audio is not stored.

Try it live · read this aloudBuongiorno, come stai?
Free · up to 8 s · audio is not stored
POST /v1/assessExample it-IT · 1.8 s audio · 1 credit
bwondʒornoBuongiorno70komecome93staistai91
85
Overallout of 100
Pronunciation84
Completeness100
Sound detailBuongiornobwonˈdʒorno
SoundExpectedHeardScore
dʒdʒdʒ93OK
ooo90OK
rrɹ56Review
nnn95OK

The “r” in “Buongiorno” was not tapped — the rest of the word is solid.

Example result · record yourself to see a real oneFull JSON reference ↗

Until you record, the card shows an illustrative example. The demo scores one fixed sentence; with an API key you send any Italian text.

Italian rewards learners who get the double consonants right and punishes the ones who do not — pena and penna are different words. The Italian pack models geminates as distinct expected sounds, so a single n where a double was required shows up as a substitution with a lower score. The trilled r and the seven-vowel system are handled the same way.

Dimensions

Pronunciation and completeness at sentence level, word scores and phoneme evidence. Fluency and rhythm are not scored for Italian today.

Learner errors this pack is built to show

  • Single consonants where geminates are required — pena / penna, caro / carro.
  • Trill r produced as an English approximant.
  • gl and gn produced as gl / gn clusters instead of ʎ / ɲ.
  • Diphthongised vowels — o as oʊ, e as eɪ — common for English speakers.
  • c / g before e / i produced hard instead of tʃ / dʒ.

Phoneme rows are evidence, not validated corrections: they show a learner where a word went wrong, while the word-level decision is what your app should act on.

How far to trust Italian scores

Every pack grades against native-speaker references. For English, Spanish, German and Russian we have also checked the headline score against a large set of rated learner recordings and adjusted it to match. Italian has not had that learner check yet: its scale is set from native read speech only, and the wrong-sentence check is carried over from the packs that have. In practice:

  • Native speakers score high. In our October 2026 recalibration check on 40 native read-speech clips, about nine in ten native takes scored 85 or more.
  • A different sentence scores low. Takes of the wrong sentence sat far below natives in the same check (medians in the teens).
  • The middle is not validated. We have no rated Italian learner set yet, so we do not map scores to levels or pass marks. Compare a learner with their own history, or learners with each other on the same sentence.

Fluency, rhythm, stress and intonation are not scored. The pack is labelled experimental; the methodology page explains what that means and model_version in every response names the model that scored it.

Example request

Reference text: Buongiorno come stai. Send the recording, the text and the locale; nothing else is required.

cURL
curl https://api.toneperfect.app/v1/assess \
  -H "Authorization: Bearer $TONEPERFECT_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -F "audio=@take.wav" \
  -F "text=Buongiorno come stai" \
  -F "language=it-IT"

What comes back

The same structure as every other language: sentence-level scores, a words[] array with timestamps and, under each word, phones[] with expected and heard values. Abbreviated to the one flagged word:

JSON (abbreviated)
{
  "language": "it",
  "locale": "it-IT",
  "task": "scripted_speech",
  "status": "completed",
  "assessable": true,
  "model_version": "multilingual-native-v1-it-…",
  "scores": {
    "overall": 85,
    "pronunciation": 84,
    "completeness": 100
  },
  "words": [
    {
      "word": "Buongiorno",
      "expected_ipa": "bwondʒorno",
      "score": 70,
      "start_ms": 180,
      "end_ms": 560,
      "decision": "review",
      "phones": [
        {
          "phone": "dʒ",
          "expected": "dʒ",
          "heard": "dʒ",
          "score": 93,
          "diagnosis_status": "experimental"
        },
        {
          "phone": "o",
          "expected": "o",
          "heard": "o",
          "score": 90,
          "diagnosis_status": "experimental"
        },
        {
          "phone": "r",
          "expected": "r",
          "heard": "ɹ",
          "score": 56,
          "diagnosis_status": "experimental"
        },
        {
          "phone": "n",
          "expected": "n",
          "heard": "n",
          "score": 95,
          "diagnosis_status": "experimental"
        }
      ]
    }
  ],
  "billing": {
    "credits": 1,
    "balance_credits": 2499
  }
}

Reference text goes in as text: The Italian sentence the learner was asked to read, up to 320 words. See the response reference for every field.