Worked scenario
Example: pair three slide images with three scripts, preview each voice segment, then start the presentation in the intended order.
Turn numbered image and text pairs into a narrated, captioned slideshow video. Free and browser-based; translate slides into Urdu, Arabic, French and more.
Select image files and their matching text files together (you can also add each group separately). Pair them by the number at the start of the filename — 1.jpg goes with 1.txt, 2.jpg with 2.txt, and so on. Every image needs a matching text file.
When set to a language other than English, your slide text is automatically translated before it's spoken and captioned. Requires an internet connection.
For audio recording, use desktop Chrome or Edge, select this tab and enable Share audio. Some system voices cannot be captured: test a short recording before a full presentation. Microphone access is not requested.
Turn numbered image and text pairs into a narrated, captioned slideshow video. Free and browser-based; translate slides into Urdu, Arabic, French and more.
Pair numbered image and text files. Voice and audio-capture support depend on the browser and operating system.
Pair slide images with narration text, select a browser voice and present the sequence. Voice availability depends on the device.
Playback uses installed browser voices. Recording combines the slide canvas with the audio explicitly shared in the browser dialog.
Illustrative inputs for checking the method; these are not project records.
Example: pair three slide images with three scripts, preview each voice segment, then start the presentation in the intended order.
Turn numbered image and text pairs into a narrated, captioned slideshow video. Free and browser-based; translate slides into Urdu, Arabic, French and more.
Use an installed voice for the selected language. Audio recording needs desktop Chrome or Edge, HTTPS or localhost, and shared audio permission. Some system voices are not capturable; listen to a test download first.
Choose a recording mode before playback to download a WebM video. Play only produces no download. Keep the original image and text pairs to reopen your presentation; this tool has no project-save file control.
Talking Slides is a free, browser-based tool that turns a set of numbered images and matching text files into a narrated, captioned slideshow video. Load your slides, pick a voice, and play the presentation — the tool reads each slide's text aloud, scrolls the caption in sync, and can record the whole thing as a downloadable video. Talking Slides also supports multi-language text translation and voice narration: choose French, Spanish, German, Portuguese, Italian, Urdu, Arabic, Hindi, Chinese, Turkish, Russian, or Japanese from the Language menu, and your English slide text is automatically translated, then spoken aloud and captioned in that language, matched to an installed voice for that language wherever possible. Nothing is installed and nothing is uploaded: everything runs locally in your browser, aside from the text sent for translation when a non-English language is selected.
Give each image and its matching text file the same leading number, such as 1.jpg with 1.txt, and 2.jpg with 2.txt. The tool pairs them automatically by that number.
Yes. Choose a language from the Language dropdown — including French, Spanish, German, Portuguese, Italian, Urdu, Arabic, Hindi, Chinese, Turkish, Russian, or Japanese — and your English slide text is automatically translated, then spoken aloud and captioned in that language.
The Voice dropdown automatically filters to voices installed on your system that match the selected language. If none are installed, the tool warns you and falls back to your default voice, which may mispronounce the translated text.
Only if you select a voice your browser can capture. Your operating system's built-in voices play through the system directly and cannot be recorded by any browser; the exported video will be silent if one of those is selected.
No. Talking Slides runs entirely inside your web browser. Nothing is installed, and no files are uploaded to a server.