Guide
Bitrate and sample rate, explained
The short answer
Bitrate is how much data per second a lossy file is allowed, and it is what decides both quality and file size. Sample rate is how many times per second the sound was measured, and 44.1 or 48 kHz already covers everything humans hear.
By Suprabhat Updated How we write these
Bitrate: the one that matters
A 128 kbps MP3 spends 128,000 bits on every second of audio. Double it and the encoder has twice as much room to describe the sound, so it discards less — and the file is twice the size.
Where the useful ceiling sits depends on the content. For speech, 96–128 kbps is transparent to almost everyone. For music, most listeners stop hearing improvements somewhere between 192 and 256 kbps in MP3, and lower in more modern codecs like AAC or Opus.
320 kbps MP3 is not wrong, it is just usually spending 60 per cent more storage on something the listener cannot detect on the equipment they actually use.
Sample rate: mostly already solved
Sample rate sets the highest frequency that can be represented: the limit is half the sample rate. At 44.1 kHz — the CD standard — that ceiling is 22 kHz, comfortably above the roughly 20 kHz limit of good young hearing.
48 kHz is the video and broadcast standard, and worth matching if the audio is going into a video, purely to avoid a resample.
Higher rates — 96 kHz, 192 kHz — are useful while recording and processing, where the extra headroom helps some operations, and are not audibly better on playback. They mainly make files two to four times larger.
Resampling is not free: going from 44.1 to 48 kHz and back is arithmetic on every sample. Pick one and stay there.
Bit depth, the third number
Bit depth applies to uncompressed formats such as WAV and FLAC, and sets the dynamic range — the distance between the quietest and loudest sound that can be stored. 16-bit gives about 96 dB, which covers anything you will listen to. 24-bit gives more headroom for recording and editing, which is where it earns its place.
Lossy formats do not have a bit depth to set; that is what bitrate replaces.
Practical settings
Recording and editing — WAV or FLAC, 48 kHz, 24-bit.
Speech for publishing — MP3, 96–128 kbps, mono.
Music for phones and cars — MP3 at 192–256 kbps, or M4A/AAC at 128–192 kbps for the same quality in less space.
Archiving — FLAC at the source rate. Lossless, tagged, and half the size of WAV.
Audio Converter exposes all of these and keeps the source sample rate unless you tell it otherwise.
Questions
Is 320 kbps noticeably better than 192 kbps?
For most listeners on most equipment, no. The difference is real but small, and it costs 60 per cent more storage.
Should I record at 96 kHz?
Only if your workflow specifically benefits. For playback it is not audibly better, and the files are twice the size.
Does converting 128 kbps MP3 to 320 kbps improve it?
No. The detail was discarded at 128 kbps and cannot be recovered; you just get a larger file containing the same degraded audio.