Glossary
Sample rate
The number of times per second an audio signal is measured, expressed in kilohertz. Podcast production commonly records at 44.1 or 48 kHz, above what most speech models require.
Telephone audio is 8 kHz, many ASR pipelines run at 16 kHz, and production audio is captured at 44.1 or 48 kHz. Downsampling is trivial; upsampling cannot recover detail that was never captured, so archives should be delivered at their native rate. Higher rates matter most for TTS and audio generation, where output quality is bounded by input quality.
Why it matters
Native-rate masters keep every buyer option open; a downsampled copy closes some permanently.