Auto-Subtitle listens to the speech in a video and writes a timed subtitle file in SRT format — the format YouTube, Vimeo, VLC, Premiere, CapCut and most players accept. Load the .srt next to your video and the captions appear, line by line, at the right moments.
Abzar first pulls the sound track out of the video in your browser, so only the audio is sent to the speech-recognition model (Whisper), not the picture. The subtitles are written in the language that is spoken — they are not translated — and they come as a separate file, not burned into the video. Because recognition runs in the cloud, the tool needs an account, uses AI credits and is part of Abzar Plus; free accounts can use it once a day.
How to use
Sign in and add the video (an audio file works too).
Run it — the audio is extracted on your device first, then transcribed with timestamps.
Download the .srt file and read it through; fix names and unusual words in any text editor.
Upload the .srt to YouTube, or load it alongside the video in your player or editing app.
FAQ
Are the subtitles burned into the video?
No. You get a separate .srt file with timings. That keeps them editable and lets viewers switch them on or off; burning text into the picture is not something this tool does.
Which languages work, and are they translated?
The model detects the spoken language itself and handles most major languages, including English, German and Persian. Subtitles come out in that same language — there is no translation step.
How long can the video be?
The extracted audio has to fit in one request of about 25 MB, which is roughly a quarter of an hour of video. Split longer recordings first with Trim Video; the timings of each part then start again at zero.
What does it cost, and what is sent?
It currently uses 1 AI credit per run and is included in Abzar Plus; free accounts can use it once a day. Normally only the audio track is sent, over an encrypted connection, to the speech-recognition provider; the video itself stays on your device.