How to convert an audio file

Pick the format your next app needs, then choose the smallest setting that still suits the job. This guide explains the tradeoffs so you can convert once and keep a clean original.

The short answer

Open the audio converter, add your file, choose WAV, MP3 or WebM with Opus, set the output options and prepare the result. Listen to the result before saving it. The converter works in this browser tab, so your audio does not travel to a conversion server.

Conversion normally has two stages. The source codec is decoded into audio samples. Those samples are then written into a new container with the output codec and settings. The output can be smaller, easier to edit or more widely supported, but it cannot recover information already lost in the source.

Codec and container are different

A codec describes how audio samples are compressed or stored. MP3 and Opus are codecs. A container packages audio with a header and sometimes tags, artwork or other streams. WAV commonly carries PCM samples. WebM commonly carries Opus. The filename extension is only a clue, so renaming song.mp3 to song.wav does not convert anything.

When somebody asks for “WAV”, they usually mean PCM audio in a WAV container. When they ask for “WebM”, check that they also accept Opus audio. A file can have a familiar extension and still be rejected if its stream is the wrong codec or channel layout.

Choose a format by the next job

For a one-off handoff, ask what the receiving application accepts before you convert. A 320 kb/s MP3 is still the wrong answer if the upload form only accepts WAV. A WAV is still the wrong answer if a messaging service imposes a small size limit.

Lossless and lossy conversion

Lossless storage keeps enough information to reproduce the same samples. WAV PCM is uncompressed, and FLAC compresses those samples without changing them. A lossless conversion can change the container and file size while keeping the decoded samples identical.

Lossy codecs reduce data by removing information that is expected to be less noticeable. MP3 and Opus do this deliberately. Encoding a clean source once is normal. Decoding an MP3 and encoding it as another MP3 asks a second lossy encoder to work from an already reduced signal. It cannot restore the first pass and can add ringing, swishing or high-frequency loss.

Converting MP3 to WAV is therefore lossless from that point forward, but it is not a quality upgrade. Keep the original WAV, AIFF or FLAC master when you have one, and make delivery files from that master.

Sample rate: match the destination

Sample rate is the number of samples captured each second. It is written in hertz, so 44.1 kHz means 44,100 samples per second per channel.

Resampling does not add detail. Upsampling a 22.05 kHz recording to 48 kHz creates more sample positions, not new high-frequency information. Keep the source rate when there is no downstream requirement. Change it when the next app, delivery specification or file-size target gives you a reason.

Channels and bit depth

Channels describe the number of separate tracks in the file. Stereo preserves left and right differences. Mono combines the source into one channel and can almost halve a PCM file. That is useful for a single microphone or spoken recording, but it changes a stereo mix. Check for phase cancellation before collapsing a music file to mono. Duplicating mono into two stereo channels makes a larger file without creating a second perspective.

Bit depth describes the precision of each PCM sample. Sixteen-bit is a practical delivery setting and is widely supported. Twenty-four-bit gives an editor more level headroom during production. Thirty-two-bit float is useful inside some production applications, but it does not make an ordinary 16-bit or MP3 source more detailed. The converter offers 16-bit, 24-bit and 32-bit WAV output so you can match the next step.

MP3 bitrate: the size and quality control

Bitrate is the encoder's approximate budget per second. A higher value usually keeps more detail and makes a larger file. It is not a measure of the source quality, and increasing it after the source has already been compressed cannot bring lost detail back.

Use the lowest setting that meets the delivery requirement, then listen to difficult passages such as cymbals, sibilant speech and quiet fades. For repeated editing, do not use MP3 as your working master. Export a delivery MP3 once from the cleanest source you have.

Estimate the file size before converting

For uncompressed PCM, the audio payload is predictable:

sample rate × channels × bytes per sample × seconds

Five minutes of stereo 48 kHz 16-bit WAV is:

48,000 × 2 × 2 × 300 = 57,600,000 bytes, or about 57.6 MB before the small WAV header.

For a constant-bitrate MP3, use:

bitrate × seconds ÷ 8

Five minutes at 192 kb/s is:

192,000 × 300 ÷ 8 = 7,200,000 bytes, or about 7.2 MB. Variable-bitrate encoders can land above or below that estimate because quiet and complex passages receive different amounts of data.

Changing stereo to mono halves the PCM channel term. Changing 48 kHz to 44.1 kHz reduces the sample-rate term by about eight percent. Those savings alter the audio format, so preview the result and check the final file in the receiving app.

Convert with floi

1. Open the audio converter and drop in an MP3, WAV, M4A, AAC, OGG, Opus, FLAC, AIFF, CAF or WebM file.
2. Choose WAV PCM for editing, MP3 for broad compatibility, or WebM with Opus when the option is available and the destination accepts it.
3. For WAV, choose sample rate, channels and bit depth. For MP3, choose 128, 192, 256 or 320 kb/s.
4. Press Prepare conversion. The page decodes and encodes locally. It does not change the source file.
5. Use the built-in player to compare the source and result. Save only after checking the duration, channel choice and difficult parts of the audio.

The input gate is 500 MiB, but decoded audio can require several times the source file's size in memory. A long, compressed file can therefore fail on a phone even when it is below the file-size limit. A desktop encoder is a better fit for long programmes or batch work.

The generated file contains fresh audio data. Metadata, cover art and chapter tags are not copied through this decode and write path. If those fields matter, keep the source and add tags with a format-specific editor after the conversion.

Convert with FFmpeg

FFmpeg is useful when you need repeatable commands, long files or batch conversion. These examples run on your computer. They are not commands executed by floi.

ffmpeg -i input.mp3 -c:a pcm_s16le output.wav
ffmpeg -i input.wav -c:a libmp3lame -b:a 192k output.mp3
ffmpeg -i input.wav -c:a libopus -b:a 96k output.webm
ffmpeg -i input.wav -ac 1 -ar 48000 -c:a pcm_s16le output-mono.wav

The first command decodes MP3 and writes 16-bit PCM. The second makes a 192 kb/s MP3. The third makes WebM with Opus at 96 kb/s. The fourth changes both channel count and sample rate, so listen for the tradeoff.

Extracting audio from a video is a different job from converting it. This command copies an AAC stream into an M4A container without re-encoding:

ffmpeg -i clip.mp4 -vn -c:a copy audio.m4a

Use stream copy when the existing codec and container suit the destination. Replace -c:a copy with an encoder only when you actually need a different codec, bitrate or channel layout.

Use Audacity when you need an edit

Audacity is a strong choice when conversion is only one part of the job. Import the file, edit the waveform, apply fades or noise reduction, then choose File → Export Audio and select the required format. It decodes the source on import and encodes a new file on export, so exporting an MP3 from an MP3 is another lossy generation.

For a plain format change, a browser converter is quicker. For a project with several tracks, effects, automation or a saved edit, Audacity gives you the controls and project file you need. In either case, export from a WAV, AIFF or FLAC master when possible.

Common mistakes

• Renaming the extension. It changes the label, not the bytes or codec. Use an encoder.
• Upscaling bitrate. A 128 kb/s MP3 saved at 320 kb/s is a larger copy of the same reduced source.
• Resampling without a requirement. It adds a processing step and cannot create detail.
• Making music mono to save space. It can remove useful stereo information or expose phase problems.
• Saving MP3 repeatedly. Make edits from the cleanest source and encode a delivery copy once.
• Replacing the source immediately. Keep the original until the output has passed a real playback check.
• Checking only the source player. Test the saved output in the app, device or upload form that will receive it.

Which method should you use?

A safe conversion checklist

1. Keep the cleanest original you have.
2. Ask what codec, container, sample rate and channel layout the destination accepts.
3. Choose WAV or another lossless working format for editing.
4. Choose MP3 or Opus only when the size or delivery target calls for lossy compression.
5. Do not increase bitrate or sample rate expecting lost detail to return.
6. Listen to the saved output, then check its duration, size and compatibility.
7. Keep the source until the recipient or upload service accepts the result.