Upload a video or audio file. We extract the audio, transcribe Tigrinya speech, and align every word to the timeline — ready for an editor that actually understands ግዕዝ script.
We extract the audio, transcribe Tigrinya speech, and align every word to the timeline. Then you can edit and export captions.
Built specifically for Tigrinya — including ግዕዝ script handling, Tiglish input, and forced alignment that respects diacritic-bearing syllables.
FFmpeg lifts the cleanest mono 16 kHz track from MP4, MOV, MKV — no re-encoding, no quality loss.
~5sA model fine-tuned on hours of native Tigrinya speech, with diacritic-aware tokenization and code-switch handling.
~25s per minute of audioEvery word gets a precise start and end timestamp. Cue boundaries lock so edits never drift.
Frame-accurateFix mistranscriptions inline with Tiglish input. Export SRT, VTT, or TXT — drop straight into YouTube or Premiere.
SRT · VTT · TXTEach word is aligned to the audio with forced-alignment, so highlights track speech precisely. No more cues that drift two seconds late.
Fix mistranscriptions inline. Type Latin to get ግዕዝ via the built-in Tiglish input. Cue boundaries stay locked so your timestamps never drift.
Download as SRT, VTT, or plain TXT. Drop into YouTube, Premiere, Final Cut, or your captioning tool of choice — no proprietary formats.