🎬 Publish the tutorial
Record the walkthrough in HD, let Whisper caption it (because most viewers watch on mute), and compress the result until it uploads everywhere without complaint.
Sixteen tools for recording, converting, compressing, transcribing, and editing sound and video, all running in your browser. That matters double here: your footage stays private, and a two-gigabyte video starts processing instantly instead of spending an hour uploading to someone's server first.
Recording and shrinking video lead the list, the on-device AI trio (transcription, subtitles, voiceovers) follows, and the cutting-and-stitching utilities round it out. Everything is powered by FFmpeg and AI models running on your own machine.
Full screen, one window, or a tab, in HD, with system audio and mic mixed in.
Tutorials, demos, bug reports, and the meeting that should have been a video: record straight from the browser with no install, no watermark, and no time limit. Capture the whole screen, a window, or a single tab, mix system audio with your microphone, and export MP4 or WebM. The subscription recorders charge monthly for this because their servers process your video; here yours does.
Shrink any video to a sendable size. No upload, no cap, no watermark.
The phone shot four gigabytes and the form accepts one hundred megabytes. Pick a quality and resolution and FFmpeg, running in your browser, re-encodes the video into a dramatically smaller MP4. Because nothing uploads, a huge file starts compressing the second you drop it, and the result comes out clean: no watermark, no trial banner, no resolution held hostage by a pro tier.
Trim any clip into a looping GIF at the frame rate and size you choose.
GIFs still rule chats, docs, and pull requests. Trim the exact moment, balance smoothness against file size with the frame rate and dimensions, and download instantly. The reaction GIF of your coworker, the three-second product demo, the screen capture that explains the bug better than a paragraph: all thirty seconds away.
Batch convert MP3, M4A, WAV, OGG, and FLAC, or pull the audio out of a video.
The voice memo is M4A, the editor wants WAV, the podcast host wants MP3 at a specific bitrate. Convert one file or a whole batch between five formats with bitrate control, or hand it a video and extract just the soundtrack. Everything lands in a single ZIP, and the no-size-cap rule matters here: hour-long recordings convert as happily as ringtones.
Whisper transcription in your browser: audio in, editable timestamped text out.
Drop in the meeting recording or the interview and OpenAI's Whisper model, running on your own device, types it out. Click any timestamp to jump the audio, correct the words it misheard, and export TXT, Markdown, or subtitle files. Transcription services charge by the minute and keep a copy of your audio; this one is free and keeps nothing.
Auto-caption any video with Whisper, edit and style the cues, export or burn in.
Most social video plays on mute, so captions are not optional anymore. Whisper transcribes the video on your device, you tidy the cues and style the text, and then choose: export SRT or VTT files for platforms that take them, or burn the subtitles permanently into a new MP4 for the platforms that do not. A caption workflow that usually costs a subscription, free and private.
Any script becomes a natural AI voiceover: ten voices, MP3 or WAV out.
Narration for the explainer video, audio versions of your writing, or a voice for the project that does not have a narrator. Pick from ten voices, tune the speed, and download MP3 or WAV. The voices are generated on your device with no character limits, which is the part the per-word pricing pages of TTS services would rather you not know is possible.
Record from your mic with a live level meter, then download the MP3.
Voice memos, dictation, podcast takes, the pronunciation guide for your name: hit record, watch the meter confirm the mic hears you, pause and resume as needed, and play it back before downloading. Recordings are among the most personal files there are, and these never leave the device that made them.
Test the mic, camera, and speakers before the call instead of during it.
"Can you hear me now?" is not an opening line. Check the mic with a live level meter and waveform, record and replay a few seconds to hear what they will hear, preview the camera with its real resolution and frame rate, and confirm both speaker channels. Device pickers untangle the headset-versus-laptop mystery, and clear troubleshooting handles the permission prompts.
White, pink, and brown noise plus fan and ocean, synthesized live with no loops.
Focus in a loud house, sleep in a loud world. Five sounds generated in real time by your browser's audio engine, so there is no loop seam to catch your ear at 2am and nothing streams from anywhere. A tone control warms or brightens the mix, and the sleep timer fades out gently instead of stopping cold.
Cut a long video into the segments you actually need, same format out.
The two usable minutes inside the hour-long recording: mark the segments, cut, and download each piece in the original format. Lecture highlights, the good take from the third attempt, the clip worth posting from the full event: extracted without re-encoding your way through a desktop editor.
Strip the soundtrack in seconds with zero quality loss. The video is copied, not re-encoded.
Mute the clip before it goes anywhere: the background conversation, the wind, the commentary that seemed funnier at the time. The audio track is removed while the video stream is copied bit-for-bit, so it takes seconds even for big files and the picture quality is untouched. MP4, MOV, WebM, MKV, AVI, and more.
Lay music, a voiceover, or an effect onto any video, aligned on a real timeline.
The montage needs music; the demo needs narration. Drag the audio clip along a shared timeline, line it up against the audio waveform and video thumbnails so the beat drops where the scene changes, and export. Paired with the Text-to-Speech Studio, it turns a silent screen recording into a narrated walkthrough without a video editor in sight.
Stitch multiple clips into one video and download the combined file.
Three phone clips from the recital, the intro plus the main recording, the before and the after: put them in order and merge them into a single file. Simple by design, private by architecture, and free of the watermark that cloud editors stamp on free merges.
Cut MP3s into segments, with a preview before every download.
The best thirty seconds of the song, the single question from the hour-long interview, the ringtone-to-be: set the in and out points, preview the cut until it is right, and download the segment. The audio equivalent of cropping a photo, and about as fast.
Stitch audio files together with trims, fades, and crossfades, out to WAV.
The podcast intro, the interview, and the outro become one file. Trim each segment, add fade-ins and fade-outs, crossfade between pieces so the joins disappear, preview the whole sequence, and export WAV. A surprising amount of real audio production for a browser tab.
Clean hum, hiss, fan, and traffic out of a voice recording with a neural noise model.
The air conditioner you stopped hearing years ago is all over your recording. RNNoise, the neural suppressor inside major voice-chat apps, runs as 110 KB of WebAssembly in the page: hold a button to compare before and after, dial the strength so the voice stays natural, and export WAV or MP3. Interviews and voice memos never leave your device.
Drop boings, truck horns, applause, and 30+ more fun sounds onto any video and export an MP4.
The clip of the block tower falling over is one sad trombone away from comedy gold. Tap a sound to hear it, drop it at the playhead, drag its marker into place, and give each one its own volume and speed. All 37 effects are synthesized by the browser itself, the original audio can stay in the mix or go, and the export is a normal seekable MP4. Built for making kids videos ridiculous.
Rock-solid timing scheduled on the audio clock, with tap tempo, accents, and a speed trainer.
Most web metronomes fire clicks from page timers that wobble under load; this one schedules every click ahead of time on the Web Audio clock, so the beat is sample-steady from 30 to 260 BPM. Tap the tempo in, set 1 to 12 beats with clickable accent or silent dots, subdivide to triplets or sixteenths, then let the speed trainer raise the tempo automatically or use gap practice, which mutes bars to test your internal clock. Settings persist and the page works offline.
Mic tuning with a cents needle, 8 tunings from drop D to ukulele, and reference tones for tuning by ear.
Pluck a string and a steady, median-smoothed needle shows how many cents off you are; hold the green zone and the string gets a check mark. The autocorrelation detector resolves to about a cent from a bass low E up past a violin E string, the nearest string lights up automatically, A4 is adjustable for 442 orchestras, and every string button plays a Karplus-Strong pluck for tuning by ear. Mic audio is analyzed on your device, never recorded, never uploaded.
One-tap presets reshape any video for TikTok, Reels, Instagram, YouTube, X, or Pinterest.
Every platform wants a different rectangle, and this does the reframing without an upload: Fit keeps the whole frame over a blurred copy of your own footage (the broadcast-TV treatment) or a solid color, while Fill crops to the new shape with drag-to-reframe and up to 2.5x zoom. The live preview is exactly what exports, audio passes through or mutes, and the result is a clean seekable MP4 at the platform's native resolution with no watermark.
Slow motion or fast forward, downloaded as a new MP4 with zero quality loss.
Pick a speed from 0.1× to 4×, preview it live, and export. The trick is that your frames are never re-encoded: they are copied bit-for-bit and simply re-timed, so a long video finishes in seconds, the picture stays pixel-identical, and the file size barely changes. Keep the audio (stretched, with the classic slow-mo pitch drop) or strip it for a silent clip.
Turn an animated GIF into a smaller, sharper MP4, right in the browser.
GIF is a heavy, 256-color format; MP4 is a fraction of the size with full color and autoplays everywhere. The browser's native decoder reads every frame and its exact delay, then WebCodecs encodes an MP4 that plays back at the GIF's original pace. Pick a background for any transparent areas. No upload, no watermark.
The tools hand off to each other, and a few sequences replace entire editing subscriptions.
Record the walkthrough in HD, let Whisper caption it (because most viewers watch on mute), and compress the result until it uploads everywhere without complaint.
Mute the raw recording, generate a clean AI voiceover from your script, and lay it onto the timeline against the waveform. A silent screen capture becomes a produced video.
Record the takes, cut the keepers out of the rambles, and stitch them together with crossfades so the edits disappear. An episode, assembled entirely in the browser.
Slice the shareable moment out of the hour-long recording, then loop it as a GIF for the chat or compress it for the feed. The whole event, reduced to the part people will actually watch.