How to Keep Audio Quality When Converting

This guide explains how to keep quality high: which formats to use, which settings matter, and how to avoid the common mistakes that make a converted file sound worse than the original.

The Short Answer

Why Quality Drops During Conversion

There are two separate causes:

Step 1: Choose a Source That Keeps Detail

The quality of your output can't exceed the quality of your input. Before you convert, check the source:

• BESTLossless files (WAV, FLAC, ALAC, AIFF): the best starting point. Convert these directly to your target format.
• OKHigh-bitrate lossy files (320 kbps MP3, 256 kbps AAC): acceptable. Converting them to a different lossy format adds one more round of loss.
• AVOIDLow-bitrate files (128 kbps or below): the detail is already gone. Converting them to a higher bitrate doesn't restore it, and the file will just be bigger.

If you only have a compressed file, keep it. Don't convert it to a lower-quality format and then back again.

Step 2: Pick the Right Target Format

For most people, AAC at 256 kbps or MP3 with LAME VBR quality 2 gives the best balance of quality and compatibility. Use Opus when file size matters more than old-device support.

Step 3: Avoid Unnecessary Re-encoding

If the audio codec is already supported in the new container, copy it without changes:

• Changing only the container (for example, MKV to MP4 with AAC audio) doesn't touch the sound, so there is no quality loss.
• Use stream copy in ffmpeg (this works only when the audio codec is compatible with the target container):

ffmpeg -i input.mkv -c:a copy output.mp4

Re-encode only when you have to, such as when the codec isn't supported on the target device.

Step 4: Use the Right Settings

Bitrate

• Music: AAC 256 kbps, MP3 320 kbps CBR, or MP3 LAME VBR quality 2. Going above 320 kbps for MP3 brings little or no audible gain.
• Speech and podcasts: AAC 96 to 128 kbps (mono), or Opus 64 to 96 kbps (mono).
• Opus music: 128 kbps stereo is transparent for most listeners.

Variable bitrate (VBR) usually gives better quality per megabyte than constant bitrate (CBR), because it spends more bits on complex passages.

Sample rate and bit depth

• Keep the original sample rate (usually 44.1 kHz or 48 kHz). Upsampling a 44.1 kHz file to 96 kHz doesn't add detail. It only makes the file larger.
• Keep 16-bit or 24-bit depth if you're doing further editing. Reduce to 16-bit only for the final delivery copy.

Channels

Keep stereo files as stereo. Mixing stereo down to mono removes spatial information and should be done only for voice or podcast content where the difference doesn't matter.

Metadata and cover art

Most tools drop tags and artwork unless you ask them to keep them. In ffmpeg, add -map_metadata 0 and, for cover art, -map 0 with -c:v copy.

Step 5: Convert With ffmpeg (Recommended)

ffmpeg is free and gives you exact control over quality. These commands assume a single audio stream.

ffmpeg -i input.wav -c:a libmp3lame -q:a 2 -map_metadata 0 output.mp3

ffmpeg -i input.wav -c:a aac -b:a 256k -map_metadata 0 output.m4a

ffmpeg -i input.wav -c:a libopus -b:a 128k -map_metadata 0 output.opus

ffmpeg -i input.flac -c:a alac -map_metadata 0 output.m4a

The -q:a 2 option sets the highest-quality VBR level for LAME, and -b:a sets a fixed bitrate for AAC and Opus.

Step 6: Convert With a Desktop App

If you prefer a graphical tool:

• Audacity (free, Windows, Mac and Linux): open the file, choose File > Export, and pick the format and quality. Use the highest bitrate option. Audacity can also apply the 32-bit float processing path, which avoids clipping during edits.
• foobar2000 (Windows): use the Converter dialog with the LAME or AAC encoder and a high-quality preset.
• dBpoweramp (Windows): good batch conversion with high-quality AAC and MP3 encoders.

Whatever tool you use, choose the highest bitrate or quality preset, and avoid presets labeled "small" or "low" unless file size is the priority.

Step 7: Avoid Common Mistakes

• Converting MP3 to MP3 repeatedly. Each pass adds loss. Keep the original and convert from it each time.
• Using a low bitrate because the file will be small. Pick the bitrate based on the content, not on the size you want.
• Normalizing or applying effects more than once. Each processing step can add distortion or clipping, so apply volume changes and filters in a single pass when possible.
• Upsampling or adding fake high-resolution data. Bigger numbers don't mean better sound.
• Trusting the file extension. A file named .m4a can hold MP3 audio, and a .mp3 file can have a low bitrate. Check the real codec and bitrate with MediaInfo or ffprobe.
• Converting a low-quality recording and expecting it to sound better. Conversion can't restore detail that was never captured.

Step 8: Check the Result

Don't rely on the file size alone. Check the output this way:

1. 1Listen to the same passage in the original and the converted file, on the same headphones or speakers, at the same volume.
2. 2Pay attention to cymbals, sibilant speech (s and t sounds), reverb tails and quiet passages. These are where lossy encoding usually shows first.
3. 3Check the codec and bitrate with ffprobe: ffprobe -hide_banner output.m4a.
4. 4Compare the duration and channel count to confirm nothing was cut or mixed down by accident.

If you need a more rigorous test, an ABX test (in foobar2000 or similar tools) lets you check whether you can really hear the difference.

Quick Settings Reference

Frequently Asked Questions