Trim FLAC audio files online with lossless quality. Set start time and duration, adjust compression level from 1 to 12.
Process files in seconds with our optimized servers
Set exact start and end points with frame accuracy
Maintain original quality with smart re-encoding
00:02:30.500 with Duration 00:03:00.000 extracts a 3-minute song starting 2 min 30.5 sec in. Use the original FLAC length minus your start point if you want the file from a marker to the end.FLAC (Free Lossless Audio Codec) is a royalty-free, open-source lossless format released in July 2001. Its container is frame-based: every FLAC frame carries its own header and is independently decodable, which is why a properly written trim can preserve bit-perfect audio between trim points without re-encoding the entire stream. That matters because most "online FLAC cutters" actually export the result as a high-bitrate MP3 — Flixier's own help text admits this — silently turning a lossless source into a lossy file.
| Property | FLAC | WAV | ALAC | MP3 |
|---|---|---|---|---|
| Lossless | Yes | Yes (PCM) | Yes | No |
| Typical size (3-min track @ CD quality) | ~20-25 MB | ~30 MB | ~22-27 MB | ~3-7 MB |
| Compression | Yes (~50-60% of WAV) | None | Yes (~60-70% of WAV) | Yes (lossy) |
| Bit depths | 4-32 bits | 8, 16, 24, 32 | 16, 24 | N/A |
| Max channels | 8 | Many | 8 | 2 |
| Max sample rate | 1,048,575 Hz (spec) | 4 GHz (spec) | 384 kHz | 48 kHz |
| Embedded metadata | Vorbis comments + cover art | RIFF chunks (limited) | iTunes-style | ID3v1/v2 |
| Royalty-free, open | Yes | Yes | Yes (Apple open-sourced 2011) | Yes (patents expired 2017) |
| Native iOS | iOS 11+ | All versions | All versions | All versions |
| Native Windows Explorer | Win 10+ | All versions | Limited | All versions |
Sources: RFC 9639 (FLAC spec), Apple ALAC source release (2011), Wikipedia FLAC and ALAC articles.
| Level | Relative Encode Time | Size vs Level 8 | Audio Quality | Typical Use |
|---|---|---|---|---|
| 0 | ~1x (fastest) | +4-6% larger | Bit-identical | Real-time capture |
| 5 (libFLAC default) | ~3x | +1-2% larger | Bit-identical | General ripping (EAC, dBpoweramp) |
| 8 (FLAC preset --best) | ~10x | Baseline | Bit-identical | Most archival rips |
| 12 (xconvert default) | ~20x | -0.5 to -1.5% smaller | Bit-identical | Maximum-density archive |
All levels decode to exactly the same PCM samples — the MD5 of the unencoded audio stays identical across levels. Higher levels just spend more CPU finding better prediction coefficients to shave a few bytes off the file. Level 8 has been the long-standing reference recommendation; levels 9-12 yield diminishing returns (often under 1%).
No. FLAC is a lossless codec — the audio data between your trim points is the same PCM samples as the source. The compression algorithm doesn't approximate or discard any information; it stores the exact original waveform. The only thing that changes when you re-encode the trimmed segment is which prediction coefficients libFLAC picks (a function of compression level), and those are mathematically reversible. If you decode the trimmed FLAC to WAV, the samples inside the kept range are bit-identical to the source.
A lot of browser-based audio editors are built on a generic Web Audio pipeline that decodes everything to a PCM AudioBuffer and then encodes the export through a built-in MP3 or AAC encoder. That's fine for a podcast cut but destroys FLAC's whole point. Flixier's own FAQ admits "Flixier exports all audio files as high-bitrate MP3s, so your files will lose some fidelity." This tool keeps the codec at FLAC end-to-end, so what you download is still a lossless FLAC.
Per RFC 9639 (the current FLAC specification, published 2024), FLAC supports 4 to 32 bits per sample and sample rates from 1 Hz to 1,048,575 Hz. In practice the common values you'll see are 16-bit/44.1 kHz (CD), 24-bit/48 kHz (studio), 24-bit/96 kHz and 24-bit/192 kHz (hi-res). FLAC supports up to 8 channels (mono through 7.1).
Yes. Leave Audio Sample Rate at "Unchanged" (the default) and Audio Channel at "Original." The output FLAC will carry the same 24-bit depth and 96 kHz sample rate as the source. Don't change the sample rate unless you specifically need CD-rate output — downsampling from 96 kHz to 44.1 kHz applies a resampling filter and is not reversible, even though both ends are still "lossless" in the codec sense.
FLAC compression efficiency varies across the track. A quiet passage compresses much better than a loud, complex one because the linear-prediction residuals are smaller. If you trim out a dense orchestral climax and keep a sparse intro, the kept portion's per-second size will be lower than the file's average. The MD5 of the unencoded audio in the FLAC streaminfo header is also recomputed for the new shorter stream, so the headers differ too.
Always trim in FLAC first, then convert to MP3 if you need the lossy copy. Trimming a FLAC is lossless; trimming an MP3 either re-encodes (generation loss) or leaves you with imperfect frame-aligned cuts (since MP3 frames overlap via the bit reservoir). Cutting first in the lossless master and encoding once at the end is the cleanest workflow. See FLAC to MP3 for the encoding step or FLAC to WAV if you need uncompressed PCM.
FLAC stores metadata in separate blocks (Vorbis comment block for tags, PICTURE block for embedded cover art, APPLICATION blocks for things like ReplayGain). A trim operation that just rewrites the audio frames typically preserves these metadata blocks. Track-specific tags like TRACKNUMBER, TITLE, and ARTIST stay; ReplayGain values, however, are tied to the original audio's loudness and should be recomputed after a non-trivial trim (most music players will do this on import).
The inputs accept millisecond precision in HH:MM:SS.sss format, which is more than enough for music editing — one millisecond at 44.1 kHz is about 44 samples. The actual cut snaps to the nearest FLAC frame boundary internally (typically every 4608 or 16384 samples at the default block sizes), so you may be off from the requested point by up to ~0.1 to 0.4 seconds in the absolute worst case. For sample-accurate edits on the order of single samples, work in WAV/AIFF instead.
Not in one operation — the CUE file isn't read here. Workflow: open the CUE in a text editor to read each track's INDEX 01 timestamps, then queue the same FLAC multiple times with different Start Time and Duration values (one per track). Alternatively, use a dedicated CUE-aware tool like CUETools locally and bring the per-track FLACs here only to refine trim points or change compression level. For batch audio work on other formats see Audio Cutter, Trim WAV, or Trim MP3.
Identical lossless audio can sound different on playback if the players use different resamplers, ReplayGain settings, or output bit-depth handling. Foobar2000 by default applies ReplayGain track gain if tags are present; VLC by default does not. If you've trimmed away the loudness profile the tags were computed against, the gain values are stale. Strip or recompute ReplayGain, or disable it in your player, and the two files will sound the same.