SOLIPSIUMAlpha
SOLIPSIUM Super resolution audio engine
Free audio file info
The report lists the file's format, length and bitrate; each sound track's codec, sample rate, bit depth and channels; each picture track's size and frame rate; and the tags. You can copy it as text. Your browser reads only the headers, without decoding the sound or the picture, so a video of several gigabytes is read a slice at a time.
Drop an audio or video file
Read in your browser. It stays on your device until you ask for a repair.
MP4, MOV, MKV, WebM, MP3, WAV, FLAC, Ogg, AAC and MPEG-TS files of any size.
This page could not read the container, so your browser's decoder opened the sound instead. That gives the channels and the length. A decoder converts the sound to its own sample rate, so the file's rate is not known here.
The audio spectrum check draws the sound itself and shows where it stops, which the bitrate alone cannot tell you.
For a spoken recording, Solipsium can run its check in your browser. When it finds clipping, dropouts, hum or background noise, it offers to repair the first 20 seconds, which you then hear beside the original.
Solipsium's check in your browser takes recordings of up to five minutes. With a free account, the editor scans recordings up to 30 minutes.
Extract audio from video saves a video's sound as MP3 or WAV in your browser, then checks it for noise and hum.
After signing up, drop the same file into the editor to repair all of it.
Then drop the same file into the editor. Scans and short repairs are free.
What the numbers mean
The report comes from the file's own headers and the headers of its packets. When the container cannot be read, your browser's decoder opens the sound and gives the channels and the length.
The container is the file format, such as MP4, MOV, WAV or Ogg, and it holds one or more tracks. The codec is how each track was stored. AAC, MP3, Opus and Vorbis leave out sound you are unlikely to hear, FLAC packs every sample into less space, and PCM keeps every sample as it is. The codec string is the exact name a browser or player checks before it plays the track.
The sample rate is how many times a second the sound was measured: 44.1 kHz on a CD, 48 kHz in most video, and 8 or 16 kHz in many phone calls. A file can hold sound only up to half its sample rate. Bit depth is how finely each sample is stored. PCM and FLAC sound have one; MP3, AAC, Opus and Vorbis do not.
A track's average bitrate is how much data it uses each second, read from its packets. The whole file bitrate is the file's size over its length, so it also counts cover pictures and the container's own data. Converting a small file to a larger format raises the bitrate and keeps every loss the small file had.
For a video, the size is the picture as a player shows it, after any rotation. The frame rate comes from the usual spacing between the frames. HDR means the picture uses a wider range of brightness and colour than ordinary video.
Tags are the words stored with the sound: title, artist, album, track number, date, genre and comment. The report says whether the file holds a cover picture, and does not show the picture itself.
Questions
Only between files saved with the same codec. The bitrate says how much data the encoder kept each second. A WAV made from an MP3 has a far higher bitrate than the MP3 and still has every loss the MP3 had. The audio spectrum check shows where the sound stops, which says more about what was kept.
The file holds none. Many recorders and converters save no tags, and WAV files often have none.
MP3, AAC, Opus and Vorbis store no bit depth, so the report has none to show. The page reads it from PCM sound, as in most WAV files, and from FLAC.
In its own group under the picture track, with its codec, sample rate, channels and bitrate. A video with a second sound track, such as another language, lists both.
The page reads the headers of up to 2,000 packets of each sound track, and up to 600 of a picture track, and gives their average. For a file whose bitrate varies, that is the average over the part read, so it can differ a little from the average over the whole file.
Not to read it. Your browser reads the file on this device, and Solipsium's check of a spoken recording runs in your browser too. Audio is sent only if you press Repair the first 20 seconds, which this page offers when the check finds clipping, dropouts, hum or background noise. Then only the recording's first 20 seconds leave this browser; Solipsium repairs them on its server and deletes them within an hour.
Drop a recording and Solipsium checks it in your browser, then offers to repair its first 20 seconds for you to compare with the original.
The check and the 20-second try need no account.