lingolix

Python client for the Lingolix pronunciation scoring API. Send a recording, get back per-syllable accuracy with expected vs. detected IPA, pitch and timing — in 17 languages.

Get an API key at lingolix.com/api-portal, then set LINGOLIX_API_KEY in your environment. Keep it server-side — never ship it to a browser.

Install

pip install lingolix      # or: uv add lingolix

Usage

from lingolix import Lingolix

with Lingolix() as client:                       # key from $LINGOLIX_API_KEY
    r = client.check("audio.wav", sentence="Guten Morgen", language="de")

    print(r.overall, r.accuracy, r.fluency, r.speaking_rate)
    for word in r.words:
        for s in word.syllables:
            print(s.text, s.expected_ipa, "→", s.detected_ipa, s.pitch, round(s.accuracy, 2))

Async is the same class with an Async prefix:

from lingolix import AsyncLingolix

async with AsyncLingolix() as client:
    r = await client.check(audio_bytes, sentence="こんにちは", language="ja")

check() takes a path, bytes, or any file-like object. Drop sentence and the audio is transcribed first, then scored against its own transcription — useful for free-form speech.

Result shape

pitch is one of "high", "flat", "low", "unknown".

Errors

Every failure is a subclass of LingolixError carrying .status and .detail.

QuotaExceededError also carries plan, remaining_minutes, quota_percentage and upgrade_url:

from lingolix import Lingolix, QuotaExceededError

try:
    r = Lingolix().check("audio.wav", language="en")
except QuotaExceededError as e:
    print(f"{e.plan} plan is out of minutes — upgrade at {e.upgrade_url}")

Languages

bg cs de el en es fi fr hu it ja ko nl pl pt sv uk — the default is ja.

Audio

Anything ffmpeg can read: WAV, MP3, M4A, OGG/Opus, FLAC, WebM. Uploads are capped at 10 MiB, checked client-side before sending. Prefer a compressed format — a single utterance is well under 1 MiB as MP3 or Opus.

Links

• TypeScript client
• API documentation
• Pricing
• Source and issues

MIT licensed.