Using the API

Text parsing

The same engine, from text: testing, chat, or your own speech-to-text.

POST /api/v1/text/parse takes JSON with text and the same schema, schemaVersion, locale, sessionId and context. After the transcript, voice and text converge: the same extraction, validation and result.

curl

curl -X POST https://api.example.com/api/v1/text/parse \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "بدي ٢ شاورما ووجبة زنجر والتوصيل لخلدا", "schema": "restaurant_order"}'

Use it to test a schema without recording, for chat channels, or when you already have a transcript. transcription.confidence is 0 for text, and recognition confidence does not lower field confidence.