Voice Analysis API
The Vocametrix API gives your application the acoustic voice measures speech-language pathologists use: AVQI, ABI, CPP, jitter, shimmer, HNR, DSI, CSID, GNE, H1-H2, formants, the S/Z ratio and more. The core measures are computed with Praat 6.4.60 and the same settings as the Vocametrix website, so your numbers match the ones clinicians see there.
Read the API documentation →Two calls
Upload a recording to get a file ID, then ask for a measure on that ID. Every call carries your key in the X-API-Key header. Files can be up to 50 MB.
# 1. Upload a sustained vowel
curl -X POST https://platform.vocametrix.com/api/assignFileId \
-H "X-API-Key: $VOCAMETRIX_API_KEY" -F "audio=@vowel_a.wav"
# {"fileId": "..."}
# 2. Compute jitter and shimmer on it
curl "https://platform.vocametrix.com/api/jitter-shimmer?svFileId=$FILE_ID" \
-H "X-API-Key: $VOCAMETRIX_API_KEY"Available measures
Most measures take a sustained vowel (svFileId): calculate-cpp, jitter-shimmer, calculate-hnr-multiband (with optional age and gender), calculate-gne, calculate-h1-h2, calculate-dsi, calculate-formant-statistics, calculate-voice-dynamics and calculate-spectral-advanced. calculate-avqi (versions 02.03 and 03.01) and calculate-abi also take connected speech (csFileId), and calculate-csid takes connected speech only. POST /api/voice-assessment runs a combined assessment from a vowel and a speech sample in one call.
How the measures are computed
The methods page lists the Praat commands, the settings and the published references behind each measure. The jitter and shimmer response also reports the recording quality, so you can reject a take that is too short or too noisy before trusting the numbers.
Pricing
You pay for the seconds of audio each call analyses, with no rounding up. Credits come in one-time packs, not a subscription: Developer Pack 40min at 5 EUR and Developer Pack 400min at 15 EUR. Creating an account is free, but there is no free tier on the API: the first call already needs a pack.
SDKs and assistants
Official clients wrap the same calls: npm install vocametrix for JavaScript and TypeScript, pip install vocametrix for Python. The measures are also available to Claude and other assistants through the Vocametrix MCP server.
Related
Score pronunciation against a reference text with the same key, or use the measures without code on the website.
Frequently asked questions
Is there a free tier?
No. The account is free, and the website comes with a 7-day trial, but API calls are paid from a credit pack from the first one.
Do I need Praat installed?
No. Praat runs on the Vocametrix server; you only send audio and read JSON.
Does the website subscription include API access?
No. The website subscription and the API credit packs are separate products.