What is bitrate? kbps and Mbps explained

Bitrate is how much data a file uses to encode one second of audio or video, usually measured in kilobits per second (kbps) for audio and megabits per second (Mbps) for video. It's a direct measure of how much information is packed into each second of playback, not a percentage or a quality score. A 320 kbps MP3 encodes twice as much data per second as a 160 kbps MP3 of the same song. More data per second generally means more of the original signal survives compression, which is why bitrate is the main dial that trades quality against file size in any lossy audio or video format.

Audio bitrate typically runs in the tens to low hundreds of kbps: MP3s commonly range from 128 to 320 kbps, AAC often sounds comparable at slightly lower numbers, and lossless formats like FLAC or WAV can run over 1,000 kbps because nothing is discarded. Video bitrate operates on a completely different scale, in Mbps, because a video frame carries far more information than an audio sample. A 1080p video might sit anywhere from 3 to 10+ Mbps depending on motion and detail, while 4K commonly needs 35-45 Mbps or more. The units differ because the amount of raw data differs by orders of magnitude, not because the underlying concept changes.

CBR (constant bitrate) holds the exact same bitrate throughout the whole file, which makes file size and streaming bandwidth perfectly predictable but wastes data on simple passages (silence, a static shot) while sometimes starving complex ones (a busy mix, fast motion). VBR (variable bitrate) raises and lowers the bitrate moment to moment, spending more data where the content is complex and less where it's simple, so it generally delivers better quality per byte at the same average bitrate. The tradeoff is that VBR's final file size and, for streaming, its bandwidth demand aren't fixed in advance the way CBR's are. Most modern encoders default to VBR for storage and CBR (or a bitrate ceiling) for live streaming, where predictable bandwidth matters more than squeezing out the last bit of quality.

Bitrate controls the tradeoff between file size and how closely the compressed file matches the original, uncompressed signal. Raising it lets the encoder keep more detail, but the gains shrink the higher you go: doubling an already-high bitrate rarely produces an audible or visible difference, since there's only so much information in the source to preserve. That's why encoding guides converge on "good enough" ranges rather than "more is always better" advice. Below a certain point, though, quality drops off fast and becomes obvious: muffled highs and swirly artifacts in audio, blocky macroblocks and smeared motion in video. Bitrate is a dial with real diminishing returns at the top and a real cliff at the bottom.

pixeltools' audio compressor exposes a single quality slider from 10 to 100, and never shows a kbps number at all. Under the hood, it maps that slider straight to an MP3 bitrate with bitrate_kbps = round(64 + ((quality - 10) / 90) * (320 - 64)), which spans 64 kbps at the low end to 320 kbps at the high end, and hands that number straight to ffmpeg as -vn -acodec libmp3lame -b:a {bitrate}k. A source-code comment explains the reasoning directly: bitrate is an internal encoding detail, not something a casual user needs to interpret, so the UI labels the result "High quality," "Balanced," or "Smaller file" instead. That's the practical lesson bitrate teaches: the number itself is a means to an end, and once you know what outcome you want, the raw kbps value is mostly useful to encoders and power users, not to someone who just wants a smaller file that still sounds fine.

For music you want to keep long-term, 256-320 kbps MP3 or AAC is effectively transparent to most listeners; anything you're archiving losslessly should stay in FLAC or WAV instead of picking a bitrate at all. For spoken-word audio like podcasts or voice memos, 96-128 kbps (even mono) is plenty since speech carries far less high-frequency detail than music. For video, match the platform: YouTube and most streaming services publish recommended bitrate tables by resolution and frame rate, and using their numbers beats guessing. When in doubt, use VBR at a quality target rather than hand-picking a CBR number. It's a big part of why tools like pixeltools' audio and video compressors ask what outcome you want instead of asking for a bitrate directly.