Persian speech-to-text API (فارسی)

Transcribe Persian with one request

Set language to fa and start a job. Poll GET /transcribe/{jobId} for the result, or pass a webhook_url.

curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/interview-fa.m4a", "model": "oracle", "language": "fa", "include_diarization": true, "output_format": "json"}'

Translate Persian audio into English in the same job (Oracle models only):

curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/interview-fa.m4a", "model": "oracle_mini", "language": "fa", "translate_to": "en", "output_format": "srt"}'

Persian at a glance

What to know about Persian transcription

• Persian is written right to left in Perso-Arabic script. Display it with dir="rtl" and a font that covers Persian letters (پ چ ژ گ).
• Persian uses the zero-width non-joiner (U+200C) inside words such as می‌خواهم. Keep it when you store, tokenize or compare text; stripping it changes words.
• Persian keyboards and texts mix Arabic ي/ك with Persian ی/ک. Normalize both forms before search.
• Dari (Afghanistan) has no separate code and is sent as fa; Tajik, written in Cyrillic, has its own code, tg.

Where teams use Persian transcription

• Persian-language media and diaspora broadcasting.
• Research and human-rights interviews translated with translate_to: "en".
• Lectures and community content with Persian subtitles.

Choosing a model

Start with oracle (US$1.44 per hour) for production Persian audio. Use base (US$0.48) for clean, single-speaker recordings at volume, oracle_mini (US$0.96) for the cheapest translation, and oracle_pro or oracle_max for noisy or difficult audio. KolWrite does not publish a Persian accuracy figure for this API, so test a sample of your own recordings before committing. Full comparison: models and prices.

Questions and answers