Glossary
Sample rate is how many times per second an analog audio signal is measured (sampled) when converting it to digital. Measured in hertz (Hz) or kilohertz (kHz). The Nyquist–Shannon theorem says you need a sample rate at least double the highest frequency you want to capture. Human speech tops out around 8 kHz, so a 16 kHz sample rate is plenty for transcription.
Common sample rates
8,000 Hz — telephone quality (POTS, Twilio low-tier). Limits high frequencies to 4 kHz; speech still intelligible.
16,000 Hz — Whisper's input rate. Captures up to 8 kHz, covering all human speech frequencies.
44,100 Hz — CD audio. Captures up to ~22 kHz, beyond human hearing.
48,000 Hz — film/video standard. Most modern recordings.
96,000 / 192,000 Hz — "Hi-Res Audio" — overkill for everything except mastering studios.
Why 16 kHz is the sweet spot for ASR
Whisper internally resamples every input to 16 kHz before processing. Higher sample rates aren't just unnecessary — they're discarded. So uploading a 192 kHz studio master gives you no accuracy bonus over a 16 kHz phone recording (assuming both have similar SNR). The downsample to 16 kHz is lossless for speech because no useful speech information lives above 8 kHz.
Practical implications
Don't worry about sample rate for transcription. Whisper handles anything from 8 kHz upward. The only practical concern is file size: 192 kHz audio is 12x bigger than 16 kHz audio, and you'll hit upload size limits faster. If you're recording specifically for transcription, 16 kHz mono codec at any bitrate is the most efficient choice.
In practice
Your iPhone Voice Memos app records at 44.1 kHz mono by default. Drop the .m4a in our uploader. We resample to 16 kHz before sending to Whisper. Transcription accuracy is identical to what you'd get from a 16 kHz recording. The only difference: the original file is ~3x larger.
Related terms
Further reading
Transcribe any sample rate
Get started — freeLast updated: September 20, 2026
Frequently Asked Questions
What does Sample rate mean in transcription?
How many times per second an audio signal is measured, in Hz. CD quality is 44,100 Hz; modern recordings often use 48,000 Hz. Higher sample rates capture more detail but produce larger files.
Why does Sample rate matter when choosing a transcription workflow?
Sample rate affects how you understand transcript quality, timing, compatibility, or the technology behind speech-to-text results.
Where can I apply Sample rate 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.