Ogg to FLAC converter
A project's audio folder grows into a mess of Ogg, WAV and MP3, and a release needs one shelf instead of three formats: FLAC gives stable bytes and checksums, but whatever Vorbis already discarded is stored as part of the shelf. Consolidation, not restoration.
FLAC stores lossless PCM, but the source is already decoded: any lossy source keeps its existing loss.
Processed on your device. Files aren’t sent anywhere.
Inventory the folder before choosing a shelf layout
Illustrative: an open-source project's audio folder holds 60 assets mixed across Ogg, WAV and MP3, with names that drifted over three releases. The next release needs one shelf. The layout decision comes first: shelf naming is based on the asset stem, not on whatever filename the folder currently carries.
| Found in the folder | Action | Why |
|---|---|---|
| Ogg assets | Convert to FLAC | Lossless shelf of the surviving samples |
| WAV assets | Convert to FLAC | Same samples, much smaller shelf |
| MP3 assets | Convert to FLAC | Stable bytes; lossy history noted |
| Duplicates with drifted names | Pick one stem, discard the rest | A shelf with two names per asset is not a shelf |
Building the shelf, one asset per run
One file per run, up to 60 MiB and 600 seconds of audio. A folder of sixty assets is sixty runs with discipline, and the discipline is what makes the shelf trustworthy.
- Drag the Ogg asset onto this page; the local FFmpeg worker loads on first use
- Convert, then read the reported codec, duration, sample rate and channel count
- Write the file into the shelf under the canonical stem
- Add its checksum to the shelf's checksum list as you go
Verifying the shelf
A shelf is only useful if it can be proven unchanged after a move. Two checks do that: duration comparison at conversion time, and checksums afterwards.
- The engine probes each encoded file and compares codec and duration with the source; drift beyond a small tolerance is reported as failure
- Keep a checksum list beside the shelf and re-run it after every storage migration
- Note in the shelf README which files came from lossy sources, so nobody assumes native recordings
- Read the reported sample rate rather than assuming it — encoder builds differ in the rates they accept
The size trade, stated honestly
The FLAC shelf is larger than the Ogg folder. That is the cost of consolidation: you are buying stable bytes and integrity checks, not saving space. If the goal were a lighter distribution bundle, the opposite route — FLAC to Ogg — is the right one.
Release from the shelf, not from the old folder
Once the shelf exists, every downstream job starts from it. Encoding from the mixed folder reintroduces the drift the shelf was built to remove.
- Streaming release: FLAC to Ogg, quality set per asset
- Legacy hosts that only accept MP3: FLAC to MP3, bitrate per recipient
- Importer or measurement work: FLAC to WAV for codec-free PCM
- Never encode a release from a shipped Ogg copy; that applies a second loss