Whether managing voice memos, podcast episodes, sample packs, or music tracks, audio files can quickly consume gigabytes of storage space. Uncompressed formats like WAV or high-bitrate audio take forever to upload and cannot be attached to standard emails.
By understanding how perceptual psychoacoustics and bitrate settings work, you can dramatically reduce audio file sizes while keeping dialogue crystal clear and musical instruments rich and vibrant.
How Audio Compression Works (Psychoacoustics)
Uncompressed audio formats like WAV (16-bit, 44.1 kHz PCM) store raw amplitude measurements 44,100 times every second for both left and right channels, generating 10MB of data per minute.
Perceptual audio encoders like MP3, AAC, and OGG utilize psychoacoustic models: they discard frequencies that the human ear cannot perceive or sounds masked by louder adjacent frequencies, cutting file sizes by 80% to 90% without audible degradation.
Step-by-Step: Compressing Audio in SizePress
Compress any sound recording or music track in four steps:
Upload Audio Track
Drag and drop your MP3, WAV, FLAC, AAC, or OGG file.
Select Target Bitrate
Pick 128 kbps (speech/podcasts), 192 kbps (standard music), or custom bitrate.
Client-Side Web Encoding
SizePress encodes the audio stream locally using WebAssembly encoders.
Save Smaller Audio File
Download your compact audio file ready for instant sharing or podcast hosting.
Bitrate Comparison: 128 vs 192 vs 320 kbps
Compare audio quality tiers and storage footprints:
| Bitrate | Audio Quality Level | Ideal Use Case | Approx. Size per Minute |
|---|---|---|---|
| 64 kbps (Mono) | Acceptable voice clarity | Voice notes & lectures | ~0.48 MB / min |
| 128 kbps (Stereo) | Good general clarity | Podcasts, webinars, audiobooks | ~0.96 MB / min |
| 192 kbps (Stereo) | Near CD quality | General music streaming | ~1.44 MB / min |
| 256 – 320 kbps | High-fidelity audio | DJ performances & master tracks | ~2.40 MB / min |
| Uncompressed WAV | Studio lossless original | Professional audio editing | ~10.5 MB / min |
Voice Memos vs Music Tracks: Optimal Settings
For human voice recordings (interviews, lectures, phone calls), high frequencies above 15 kHz carry little linguistic information. Encoding voice in mono at 64 kbps or 96 kbps achieves incredible 95% reductions compared to WAV while preserving 100% speech intelligibility.
Private Local Processing for Voice and Audio
Personal voice notes, confidential board meeting recordings, and unreleased music productions should never be transmitted to external servers. SizePress processes audio entirely inside your browser sandbox, ensuring zero exposure to third parties.