Change an audio file from one container and codec to another. The sample rate and channel count are carried through unchanged, so only the encoding differs.
Output formats
| Output | Codec | Discards detail | Typical use |
|---|---|---|---|
| WAV | 16-bit PCM | No | Editing, mastering, tools that need uncompressed input |
| FLAC | FLAC | No | Archiving at roughly half the size of WAV |
| MP3 | MP3 | Yes | Compatibility with older players and hardware |
| M4A | AAC | Yes | Better quality than MP3 at the same bitrate |
| Ogg | Opus | Yes | Speech and streaming at low bitrates |
Lossy conversion is one-way
MP3, AAC, and Opus encoders throw away audio the model predicts you will not miss. That decision cannot be reversed.
Converting an MP3 to WAV produces a larger file, not a better one. It contains exactly what the MP3 contained, stored without compression. Converting between two lossy formats is worse still, because the second encoder works on the first one’s output and adds its own losses on top. Go back to the original recording when you have it.
Bitrate
MP3, M4A, and Ogg expose a bitrate setting, which sets how many bits per second the encoder is allowed to spend.
128 kbpsis acceptable for speech and background music192 kbpsis a reasonable default for music256 kbpsand320 kbpsnarrow the gap with the source at a size cost
Opus does more with fewer bits than MP3, so 96 kbps Opus is roughly comparable to 128 kbps MP3. WAV and FLAC have no bitrate setting; their size follows the sample rate, bit depth, and channel count.
File size and long recordings
WAV stores every sample uncompressed, which works out to about 10 MB per minute of 44.1 kHz stereo. An hour-long recording lands past 600 MB.
Exports are written to browser disk storage when the browser provides it, which keeps memory use flat regardless of length. Without it there is a hard ceiling, and an oversized job is refused rather than crashing the tab. FLAC is the better choice for long material that needs to stay lossless.
Converting part of a file
Drag across the waveform to mark a range, then turn on Export selected range. The converter reads and encodes only that section, which is faster than converting the whole file and trimming it afterwards.