Make an audio waveform video
Give your audio a simple visual story.
Audio and artwork stay in this browser. Export up to 30 seconds at 24 fps; square, portrait or landscape; maximum 32 MiB.
Checking browser codecs…
Input audio:16 MiB,180 seconds,10,000,000 decoded frames, mono/stereo 8–96 kHz. Artwork:6 MiB PNG/RGB JPEG and 16 million pixels. Each worker job stops after 30 seconds for import/still preview or two minutes for video. Cancel releases the worker and its codec resources.
Choose audio and design the canvas
Open a local WAV, MP3, MP4, WebM, Ogg or FLAC file with one browser-decodable audio track. Use the original peak waveform and player to choose start and end times. Audio is decoded progressively inside a worker; unsupported codecs fail with a message instead of being uploaded. Set gain from −24 to 12 dB and review any clipping warning.
Choose square, portrait 9:16 or landscape 16:9 at a 360- or 720-pixel short edge. Add your own title and credit, set text direction, colors and waveform peaks or peak bars. Optional PNG or RGB JPEG artwork can contain the full image or cover the canvas with cropped edges; a dark overlay keeps text readable. Titles, credits and existing subtitle text wrap inside separate safe areas. Text that still cannot fit is refused with a clear error.
Review an accurate timeline
Render timeline frame draws the selected time in seconds into the excerpt. Its optional dashed safe-area guides appear only on that still preview. Render synchronized preview creates a real moving video of the complete excerpt at a 360-pixel short edge and 0.6 Mbps. Play or scrub that video to review the animation and audio together. A design change clears previous previews and results so they cannot be mistaken for the new settings.
Video frames are generated deterministically at 24 fps using the selected audio sample clock. The moving shapes show local peak amplitude around the current time; they are not a frequency spectrum, transcription or an exact oscillogram of every sample. Audio is linearly resampled to stereo 48 kHz and encoded at 128 kbps. Mono is duplicated into both channels, gain is applied, and out-of-range samples are clamped and counted. Final encoder packet padding can add a small duration margin; measured output duration is shown.
Add existing subtitles
Import UTF-8 SRT, WebVTT or a JSON project from Subtitle Editor & Converter. Cue times refer to the original recording; the chosen excerpt selects the matching part automatically. Up to 200 non-overlapping cues of 400 characters each are supported. Negative or reversed times, overlap, cue positioning and WebVTT metadata require editing before import. Subtitle markup is drawn literally as text. No speech recognition, translation or new captions are generated.
Create and check the completed video
Available combinations are MP4 with H.264/AAC and WebM with VP8/Opus. The browser must provide both encoders; exact dimensions, frame rate and bitrate are checked before each job. A missing combination requires an explicit available format choice. There is no remote transcoder or automatic silent fallback. Choose a video bitrate from 0.4 to 6 Mbps, create the file, then review its real playback, measured bytes, dimensions, duration and clipped sample count before downloading.
The generated example is an original four-second voice-like tone, with a stronger tone around one second. It does not record a microphone or download music. A separate silent example checks quiet animation. Try each canvas shape and compare peak bars with waveform peaks before adding artwork.
Projects, privacy and media limits
Project JSON always includes design settings, title, credit, cues and media identities. Audio and image bytes have separate inclusion checkboxes and are omitted by default. Reopening an omitted-media project asks you to reattach the original matching files; it never fetches remote media. Inclusion choices reset when a project is reopened. Projects can contain personal text or selected media, so review them before sharing. Nothing is automatically saved in local storage.
Source audio is limited to 16 MiB, 180 seconds, one or two channels, 8,000–96,000 Hz and 10,000,000 decoded frames. Artwork is limited to 6 MiB and 16 million pixels. The chosen excerpt is at most 30 seconds; the output writer has a hard 32 MiB cap. Project JSON is capped at 32 MiB. Imports and still previews stop after 30 seconds, moving preview/export after two minutes. Cancel terminates the worker and its codec resources; originals remain unchanged.
Bundled Noto fonts support Latin, Arabic, Devanagari and selected CJK text. Other glyphs can depend on browser fallback. Noto Sans font license, CJK font license and media library notices are supplied locally. The media format guide explains why browser codec support varies.
Can this post videos or provide music rights?
Download the file and upload it yourself. This tool does not search for music, acquire rights, post to social accounts, certify platform compatibility or transcribe speech. Use audio and artwork you are entitled to share, and check playback in the destination application.
Stars and feedback
Loading…
What should we improve next?
Help shape the next update to Audio Waveform Video Maker. Tell us what would make it better for you.
Prefer email? feedback@tooltulip.com
Tool reference #217