Assess Pronunciation

Assess English pronunciation quality from audio. Scores pronunciation at four levels: overall, sentence, word, and phoneme. Each score is 0-100. Phonemes are r

liveverifiedunknown~5s typical

mcpprobedbrainiallpronunciation

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-com-brainiall-pronunciation /assess_pronunciation for me.

Input

Field	Type	Description
audio_base64*	string	Base64-encoded audio data. Supports WAV, MP3, OGG, and WebM formats.
text*	string	The reference English text that the speaker was expected to read aloud.
audio_format	string	Audio format hint — one of 'wav', 'mp3', 'ogg', 'webm'. default "wav"

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-com-brainiall-pronunciation","endpoint":"/assess_pronunciation","input":{"audio_base64":"...","text":"...","audio_format":"wav"}}'

cli

npx cracked-ai run -p mcp-com-brainiall-pronunciation -e /assess_pronunciation -i '{"audio_base64":"...","text":"...","audio_format":"wav"}'

mcp

run_tool({ provider: "mcp-com-brainiall-pronunciation", endpoint: "/assess_pronunciation", input: {"audio_base64":"...","text":"...","audio_format":"wav"} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.