Audio file info: what is really in the file.
Drop any audio file to see its format, sample rate, bit depth, bitrate and length, next to what the samples actually hold: the loudness in LUFS, the true peak, any clipping, the bits really in use and where the top end stops. Several files at once, copied as text or saved as a spreadsheet. Nothing is uploaded.
drop audio files
MP3, WAV, FLAC, M4A, AIFF, OGG, Opus, voice memos, or the sound of a video. Up to 20 at once.
At a glance
| From the header | Format and codec, sample rate, channels, bit depth, duration; for MP3, CBR or VBR, the bitrate, the encoder and the channel mode |
|---|---|
| From the samples | Integrated loudness (LUFS), true peak, sample peak, clipping, DC offset, silence at each end, the bits really in use, where the top end stops |
| Measured against | ffprobe for the format and length, ffmpeg's EBU R128 meter for loudness (within 0.1 LU) and true peak |
| Out | Copy as text, or a CSV with a row per file |
| Uploads your files | No |
What each line means
Sample rate is how many times a second the sound was measured: 44.1 kHz for CD and most music, 48 kHz for video and most podcasts, 96 kHz and up for hi-res. It is read from the file and the file is decoded at that rate, not at your sound card's, so a 44.1 kHz file is never reported as 48. Bit depth is how finely each measurement is stored: 16 bits for CD, 24 for studio work, 32-bit float for files that will be processed further. Lossy formats such as MP3 and AAC have no fixed depth. Next to the depth the page reports how many bits the samples really use, which catches a 16-bit recording dressed up as 24-bit.
Bitrate is how much data a second of sound takes. For an MP3 the page reads the first frame and the Xing or Info frame that encoders add, so it can say whether the file is constant bitrate (CBR) or variable (VBR), which encoder wrote it (LAME stamps its version), and whether it is stereo or joint stereo. For VBR and for every other format it reports the average, the file's size over its length.
Loudness is the integrated loudness in LUFS, the measure Spotify, Apple Music, YouTube and podcast platforms use to turn tracks up or down: Spotify and YouTube play at about -14 LUFS, Apple Podcasts asks for -16. It is the ITU-R BS.1770 measure, computed at 48 kHz where the standard defines its filters, and the check behind this page holds it within 0.1 LU of ffmpeg's EBU R128 meter. True peak is the highest point of the waveform between the samples, in dBTP, which is what clips when the file is later converted; distributors usually ask for -1 dBTP or lower. Sample peak is the highest sample itself.
Clipping counts runs of three or more samples stuck at full scale, the flat-topped waveform of a recording that was too loud; a single full-scale sample is not counted. DC offset is how far the waveform's average sits from zero, which a faulty interface or a cheap phone microphone can leave; above a few tenths of a percent it costs headroom and can click at edits. Silence is how long the file stays under -60 dBFS at the start and end, which matters for gapless albums and for podcast hosts that add their own intros. Top end says where the spectrum stops; a sharp cutoff under 20 kHz means the file was a lossy one at some point, and the fake lossless detector explains what it means for that file.
Why check a file before you send it
Distributors such as DistroKid and CD Baby, mastering engineers and podcast hosts all publish specifications, and a file that misses them gets bounced or, worse, quietly converted. Before you upload: is it the sample rate they asked for, is it 16 or 24-bit, is the true peak under -1 dBTP, is it clipping anywhere, and is it the loudness you expect? MediaInfo and ffprobe answer the header questions on a computer you can install software on; Audacity's Contrast and Loudness tools answer some of the rest one file at a time. This page puts all of it on one screen, for twenty files at once, on any device with a browser.
If a number is wrong, the fix is usually one click away: the loudness normalizer sets LUFS and true peak, the sample rate converter changes the rate, and 44.1 kHz 16-bit WAV makes the CD-standard file most distributors still ask for.
FAQ
How do I check the bitrate of an MP3?
Drop it above. The page reads the MP3's own frames: a constant-bitrate file shows its bitrate and "CBR", a variable one shows "VBR" with its average, and LAME-encoded files show the encoder.
How do I find the sample rate and bit depth of a WAV file?
Drop the WAV above. Both are read from the file's header, and the page also reports how many bits the samples really use, which catches 16-bit audio saved as 24-bit.
How do I check the LUFS of a song?
Drop it above and read the Loudness line: the integrated loudness, measured by the BS.1770 standard that streaming services use. The check behind this page holds it within 0.1 LU of ffmpeg's EBU R128 meter.
What does true peak mean?
The highest point of the waveform between the samples, which can be higher than any sample. It is what clips when the file is later encoded to MP3 or AAC; most distributors ask for -1 dBTP or lower.
Can I check many files at once?
Yes, up to 20. Each gets its own card, and "Download CSV" saves one row per file to open in Excel, Numbers or Google Sheets.
Are my files uploaded?
No. The files are read and measured in your browser tab and never leave your computer.