DRISHTI — hear what you can't see

Add your video

Choose a video file

Ujumps here. Up to 5 minutes (200 MB max).
Choose the narration language

Ljumps to the list, Vto the mic — say “Hindi please”, or answer in any language and DRISHTI recognises which one you spoke. On auto, it listens to the video instead. A silent film has no voice to listen to — choose a language for those.
Who's in it? optional

Wjumps here. Give the names and DRISHTI says “Mr Bean reaches for the kettle” instead of “a man reaches for the kettle”. Separate people with commas, and say which one is which after a dash, a colon or the word “is” — “Ritika - woman in the red saree”. That description is how DRISHTI tells people apart when there is no famous face to recognise. Names only ever come from you: DRISHTI never guesses one, and if it can't tell who's who it keeps the description instead.
Have the script? optional

Left and right arrow keys choose between uploading a file and pasting the text. Enter opens the one you chose.

Choose a PDF or photo
WRITTEN IN

Sjumps here, then ←→ switch between uploading and pasting. A screenplay, a scene, or an IMDb-style synopsis all work. DRISHTI reads it when you press Describe, and uses it only to recognise what is already on screen.
Describe it

Dstarts. ?reads these instructions aloud at any time. Mmutes the voice.

How it works

Listen. Sarvam Saaras transcribes the audio in 1.5-second chunks and finds every window where nobody is speaking — even through continuous music.
Look. A vision model watches the frames and writes down what happens, moment by moment.
Speak. Each quiet window gets one spoken line in your language via Sarvam Bulbul — measured, re-paced and fitted so it never talks over the film.
Ask. Pause any moment and ask what's happening — in any language. The answer is grounded in the paused frame and the story so far, never a moment ahead, and spoken back to you.