Compression2 min read

Audio Bitrate and File Size Explained for Voice, Music, and Attachments

Audio size is largely bitrate × time. The right bitrate depends on whether you are preserving speech, music, or a master recording.

Run the actual file processor

PDF, video and audio size targets

Download a source or run it here using the same compressor as the homepage. The table reports the bytes from your run; no output size is assumed in advance. One KB here means 1,024 bytes.

Reference run — September 9, 2026
FileInput bytesTarget bytesOutput bytes
AUDIO output288,07824,57623,921

Reference method: the PDF uses the actual PDF processor with a Node canvas adapter; video and audio use native FFmpeg with the same encoding arguments and bitrate search as the browser processor. Browser encoders can return different sizes. See method and measurements.

Three-second test tone → 24 KB

A synthetic 440 Hz test tone, not speech or music. The WAV is encoded to AAC in an M4A file. Start playback at low volume and check the full three-second duration.

Download source file

Video and audio processing loads the browser encoder on the first run. Keep this page open. A small target can reduce quality, still produce an oversized file, or be rejected; a completed run does not guarantee that the result suits your destination.

For compressed audio, file size is closely tied to bitrate and duration. A 128 kb/s stream uses about 0.96 MB per minute: 128 kilobits × 60 seconds ÷ 8 ≈ 960 kilobytes, before small container differences.

Approximate audio sizes per minute

BitrateApprox. MB/minTypical implication
64 kb/s0.48 MBCompact speech; limited music fidelity
96 kb/s0.72 MBGood speech and lightweight mixed audio
128 kb/s0.96 MBCommon general-purpose compressed audio
192 kb/s1.44 MBMore headroom for music
256 kb/s1.92 MBHigher-quality compressed listening

These are estimates. Codec, channel count, variable bitrate, metadata, and container overhead can shift the final result.

Speech and music have different needs

  • Voice notes can remain intelligible at much lower bitrates than dense music.
  • Stereo music needs more information than a mono spoken recording.
  • Background noise consumes bitrate without adding useful speech detail.
  • A clean source compresses better than a clipped or distorted recording.
  • If the audio will be edited later, preserve a high-quality master and compress only a delivery copy.

Choose a target from the duration

A 30-minute recording targeted to 10 MB has only about 44 kb/s total available. That may be workable for speech but not for music. The same 10 MB target for a three-minute clip leaves roughly ten times more bitrate per second.

Fitabyte’s browser media compressor derives an audio bitrate from the target and duration and exports a compact playable copy. If the requested target is too small for the duration, the tool stops instead of pretending a reasonable result is possible.