Glossary
Audio compression splits into two camps. Lossy formats (MP3, AAC, Opus, OGG) throw away parts of the signal that human ears barely notice, achieving 5-15× file size reduction. Lossless formats (FLAC, ALAC, WAV/PCM) preserve every bit of the original, only compressing in mathematically reversible ways. For transcription, this distinction barely matters — what matters is whether the speech is intelligible.
How lossy compression works
Lossy audio codecs use perceptual coding — they model what the human ear can and cannot distinguish, then discard the inaudible parts. They mask quiet sounds adjacent to loud ones, drop frequencies above human hearing, and quantize the remaining data into fewer bits per sample. Result: 1/10 the file size, near-imperceptible quality loss to a typical listener at 192+ kbps.
How lossless works
Lossless codecs use traditional data compression (prediction + entropy coding) on the audio waveform, finding and exploiting redundancy without discarding anything. FLAC typically achieves 50-60% of WAV size; ALAC (Apple Lossless) is similar; raw WAV/PCM is uncompressed. A bit-exact decode reproduces the original sample for sample.
Does it matter for ASR?
Almost never. Whisper and other modern ASR models work on a mel spectrogram derived from the decoded audio. Anything above 64 kbps lossy is more than enough for the spectrogram to capture all speech information. Even 32 kbps MP3 transcribes nearly as well as a 1.4 Mbps WAV file. Save lossless for archival, mastering, or audiophile listening — it doesn't help transcription.
In practice
You record an interview on a Zoom H5 at 24-bit/48 kHz WAV (lossless, ~340 MB/hour). For archive: keep the WAV. For transcription: convert to 96 kbps MP3 (~45 MB/hour) and upload that — same accuracy, 8x faster upload. We accept both; we resample everything to 16 kHz mono internally regardless.
Related terms
Transcribe any format
Get started — freeLast updated: September 20, 2026
Frequently Asked Questions
What does Lossy vs lossless audio mean in transcription?
Lossy compression (MP3, AAC, OGG) discards audio data to reduce file size. Lossless (FLAC, ALAC, WAV) preserves every bit. For transcription, lossy at 128 kbps+ is indistinguishable from lossless.
Why does Lossy vs lossless audio matter when choosing a transcription workflow?
Lossy vs lossless audio affects how you understand transcript quality, timing, compatibility, or the technology behind speech-to-text results.
Where can I apply Lossy vs lossless audio on Transcript.you?
Use the related workflow at /audio-to-text when you want to see how this glossary concept connects to an actual transcription task.
Browse the full transcription glossary or read the complete guide to AI transcription.