Speech to Text

v2

Click to upload or drag & drop

MP3, WAV, M4A, WebM, OGG (max 10 MB)

Kurdish Speech to Text — Transcribe Sorani & Kurmanji Audio

Kurdish STT transcribes spoken Kurdish into written text. It supports both Sorani (Central Kurdish, Arabic script) and Kurmanji (Northern Kurdish, Latin script), scoring 4.54 character error rate on the FLEURS Sorani benchmark — accurate Kurdish speech recognition with a free tier.

How it works

  1. Record Kurdish speech with your microphone or upload an audio file — WAV, MP3, FLAC, OGG or M4A.
  2. Pick the dialect: Sorani or Kurmanji.
  3. Get the transcript with speaker labels and word timestamps — edit it in place, copy it, or translate it to 124+ languages. Sign in with a free account to start.

Features

  • Accurate Sorani and Kurmanji recognition (FLEURS benchmark results published on the homepage)
  • Speaker diarization — see who said what
  • Word-level timestamps with karaoke-style playback highlighting
  • Live streaming transcription from the microphone
  • One-click translation of transcripts into 124+ languages
  • Free tier included; paid plans raise file-size and usage limits

Have questions? See the FAQ or plans. Need Kurdish speech generated instead? Try Kurdish text to speech. Developers can transcribe programmatically via the TTS & STT API.