Editing ID3v2, Vorbis comments, and RIFF INFO tags
Maintaining clean music libraries, podcast feeds, and sound-effect stems starts with an accurate audio metadata editor running inside browser memory. Dropping an MP3, FLAC, WAV, or M4A file reveals technical stream headers alongside eight editable fields without sending private recordings across the network.
Writing UTF-16 ID3v2.3 frames and cover art on MP3
Under the Library of Congress MP3 File Format description, MPEG-1 and MPEG-2 Layer III bitstreams often carry leading ID3v2 headers and trailing 128 bytes of ID3v1 footer data. Clicking Load sample tagged MP3 builds a 44,100 Hz Joint Stereo clip at 192 kbps to preview full tag parsing immediately.
When you click Save updated tags, the worker strips stale ID3v1 and ID3v2 blocks and writes one clean ID3v2.3 header using synchsafe 32 bits of size telemetry. Every text tag encodes as UTF-16 LE with a byte-order mark (0x01) while storing notes under an eng COMM frame.
Updating FLAC Vorbis comments and WAV RIFF chunks
FLAC streams store audio specs inside a 34 bytes long STREAMINFO block and UTF-8 KEY=value fields inside a VORBIS_COMMENT metadata block. Saving edits preserves structural blocks such as SEEKTABLE and CUESHEET, drops PADDING blocks, and copies raw compressed audio frames unmodified.
WAV files store Title, Artist, Album, Album artist, Year, Track, Genre, and Comment inside a LIST/INFO chunk padded to 2 bytes and omit non-standard id3 chunks. If your project tracks video clips alongside audio stems, inspect video container metadata and GPS in the matching inspector.
How does FFT frequency cutoff detect fake lossless audio?
Lossy encoders like low-bitrate MP3 and AAC discard ultrasonic frequencies above a steep low-pass shelf around 15 kHz to 18 kHz to save storage space. Transcoding a 128 kbps MP3 into FLAC cannot restore missing harmonics. A local 2048-point FFT exposes that artificial cutoff shelf immediately against Nyquist.
Running 2048-point Hann-windowed FFT across active frames
After inspecting container bytes, the browser decodes compressed tracks into floating-point PCM channels using the MDN BaseAudioContext decodeAudioData documentation and mixes multi-channel audio to mono. Up to 24 active frames, selected above 10% of peak RMS, receive a 2048-point Hann window before Cooley-Tukey radix-2 FFT processing.
Averaging power spectra across those active frames reduces transient noise spikes and compares upper bins against the 200 Hz to 4000 Hz core speech and music band. A sliding 280 Hz window walks downward from Nyquist until energy exceeds minus 42 dB relative to the reference band.
Reading the 64-bin spectrum graph and Nyquist shelf
The True bitrate / spectral cutoff analyzer plots 64 dB bars from -96 dB up to peak energy and marks the cutoff with a dashed red line. Tracks extending past 18 kHz or 92% of Nyquist show full-bandwidth status without an artificial compression shelf.
Lower cutoffs around 15 kHz or 16 kHz indicate lossy upscales, while synthetic test frames that fail browser decoding skip spectral graphs cleanly without blocking tag edits. Once a track passes frequency checks, open Normalize audio volume locally to match streaming loudness targets cleanly.
Audio container tag, artwork, and stripping matrix
Container formats follow different binary rules for text frames, embedded images, and chunk offsets. Reviewing container support in the Xiph.org FLAC format specification alongside MP3, WAV, and M4A behavior clarifies which tracks accept edits versus read-only inspection and lossless stripping.
| Container format | Metadata structure | Editable text tags | Embedded artwork | Lossless metadata strip |
|---|---|---|---|---|
MP3 (.mp3) |
ID3v2.2, ID3v2.3, ID3v2.4, ID3v1 | Writes UTF-16 ID3v2.3 frames | Reads & writes APIC PNG/JPEG |
Removes leading ID3v2 and 128 bytes ID3v1 |
FLAC (.flac) |
STREAMINFO, VORBIS_COMMENT, PICTURE |
Writes UTF-8 Vorbis comments | Reads & writes type 3 PICTURE |
Strips comments, padding, and pictures |
WAV (.wav) |
RIFF fmt , data, LIST/INFO |
Writes UTF-8 LIST/INFO chunks |
Not supported by RIFF INFO | Retains fmt , data, and fact only |
M4A / MP4 (.m4a) |
moov/udta/meta/ilst atoms |
Read-only inspection only | Reads covr JPEG/PNG preview |
Drops udta/meta, shifts stco/co64 |
| MPEG stream (VBR/CBR) | Xing, Info, or VBRI frame headers | Calculated via frame header scan | Uses container ID3v2 tags | Leaves compressed MPEG frames untouched |
Inspecting Xing and Fraunhofer VBRI headers distinguishes genuine VBR encodes from fixed-rate CBR files across 48 sampled frames. After verifying clean metadata and genuine high-frequency bandwidth on a favorite song snippet, send the audio straight into custom iPhone .m4r ringtone creation for mobile alerts.
Tip: Run a quick frequency cutoff check before archiving “lossless” FLAC downloads. A sharp shelf near 16 kHz combined with a high lossless bitrate flag proves the file came from a lossy MP3 transcode.
Replacing cover art and stripping private tags locally
Music distributors, podcasters, and voice actors regularly swap album thumbnails or wipe private session notes before sending files externally. Processing every header inside browser memory ensures unreleased demos, client voiceovers, and personal metadata never leave your workstation.
Swapping PNG or JPEG artwork without altering audio frames
Click Replace cover art to attach a PNG or JPEG image verified directly by magic bytes (89 50 4E 47 for PNG, FF D8 for JPEG). On MP3 files, artwork writes into an ID3v2.3 APIC Front Cover (0x03) frame while preserving compressed audio payloads intact.
On FLAC tracks, buildFlacPictureBlock reads PNG dimensions directly from image bytes so players display accurate pixel widths and bit depths inside the PICTURE block. Click Remove cover art anytime to detach embedded graphics before clicking Save updated tags and downloading -tagged output.
Wiping ID3, Vorbis, and M4A metadata inside a Web Worker
All binary parsing, container rebuilding, and 2048-point FFT calculations run off the main thread via the MDN Web Workers API reference. Clicking Strip all metadata removes Comment, Year, encoder strings, and cover bytes, reporting exact bytes shaved when saving the -clean file.
On M4A and MP4 tracks, stripM4aMetadata removes udta and meta atoms, shrinks the moov header, and updates every 32 bits stco and 64 bits co64 offset so audio stays playable. If your clip sits inside a video container instead, silence sensitive recordings before sharing clips.