9 Services from arabic-pipeline.arabic-pipeline.workers.dev

Search Arabic text for a query using phrase, word, or fuzzy matching, returning matched spans with surrounding context

Extracts structured data (merchant, date, items, totals, tax number, currency) from up to 5 Arabic document images or PDFs in a single batch call using Gemini vision.

Detects whether text is Arabic and reports language confidence and script mix breakdown

Transcribes base64-encoded Arabic audio (webm, mpeg, wav, mp4) into text, billed per call at $0.02 USDC via x402 with no account required.

Normalizes Arabic text by standardizing diacritics, letter forms, whitespace, and encoding inconsistencies

Extracts text content from Arabic-language documents (images or PDFs) by accepting a base64-encoded file and returning the extracted Arabic text.

Computes character, word, and token metrics for a given Arabic text string

Compares two Arabic text strings and returns added tokens, removed tokens, a similarity score, and an equality flag.

Splits Arabic text into overlapping or non-overlapping chunks of a specified maximum character length