Persian-first speech-to-text for voice notes, meetings, classes, and video—with careful attention to transcription quality.
Built by SeyedAli Emami
A Persian-first speech-to-text system
Avangaar turns audio, voice messages, meetings, classes, and video into text. I built it after a serious R&D effort to find a speech-to-text service I could trust, with a particular focus on Persian transcription quality.Around four years ago, I researched and tested nearly every API and service I could find in the speech-to-text space. Sonic stood out for its work on Persian audio. That research became the starting point for Avangaar.From the systems I tested, Avangaar is among the most accurate for Persian and English audio. I have strong confidence in its output, especially for Persian—but the best way to judge it is to try a real recording.
Made for real recordings
Use Avangaar to transcribe:
audio files and voice messages
meetings and classes
video recordings
Persian and English speech
The interface keeps uploads, processing status, transcription history, and completed text together. Live transcription and speaker-separated transcripts are also part of the workflow.
Try the demo
AVANGAAR / DEMOAvangaar live transcription demo
00:00 / 00:00
Pricing
Transcription currently costs 220,000 toman per hour of audio. Send me an audio file, voice message, meeting, class, or video, and I will return the complete transcript.
