Sample rate: telephone audio uses 8 kHz, CDs use 44.1 kHz

Sample rate is how many times per second a digital recording measures the sound wave, written in hertz or kilohertz. Uploads can arrive at any sample rate.

Sample rate is how many times per second a digital recording measures the sound wave, written in hertz or kilohertz. The highest frequency it can represent is half the sample rate (the Nyquist frequency). Telephone audio uses 8 kHz, CDs use 44.1 kHz, and 16 kHz is common for wideband speech.

Also searched as "sampling rate", "sampling frequency" and "khz".

What the sources say about sample rate

  • Wikipedia, Sampling (signal processing). Sampling rate is the number of samples per second; Nyquist frequency is half of it; 44.1 kHz is the audio CD standard; 8 kHz is used by nearly all telephony (most speech energy is between 100 Hz and 4 kHz); 16 kHz extends narrowband to wideband and is used in most modern VoIP products (flagged on that page as a questionable source). Read on 22 October 2026.

Sample rate in Autrans

Uploads can arrive at any sample rate. The engine decodes everything to 16 kHz mono for recognition, and says so in the recorder code: recording a speech take at a higher quality than that buys bytes, not accuracy. When Autrans reads duration from an Ogg file it uses the file's own sample rate, because dividing a 44.1 kHz file by 48,000 comes out 8 in 100 short.

A worked example of sample rate

At 16,000 samples per second the highest frequency that can be represented is 16,000 / 2 = 8,000 Hz. At 8 kHz telephone rate it is 4,000 Hz. Duration check: a 44.1 kHz Ogg with 441,000 samples lasts 10.0 s; divided by 48,000 it would read 9.19 s, about 8 in 100 short.

Terms next to sample rate

  • Bitrate: Bitrate is the number of bits used to store or send each second of audio, usually written in kilobits per second (kbit/s).
  • Bitrate vs sample rate: Sample rate is how many measurements of the sound wave are taken each second (Hz).
  • Opus codec: Opus is an open audio codec standardised by the IETF in RFC 6716 for interactive speech and music: calls, video conferencing, in-game chat.
  • Lossless audio: Lossless audio is compressed or stored so the original can be reconstructed exactly, bit for bit.

Questions people ask

What does sample rate mean?

Sample rate is how many times per second a digital recording measures the sound wave, written in hertz or kilohertz. The highest frequency it can represent is half the sample rate (the Nyquist frequency).

How does sample rate show up in Autrans?

Uploads can arrive at any sample rate. The engine decodes everything to 16 kHz mono for recognition, and says so in the recorder code: recording a speech take at a higher quality than that buys bytes, not accuracy.

What is the difference between bitrate and sample rate?

Sample rate is how many measurements of the sound wave are taken each second (Hz). The full answer, with sources, is on the page for Bitrate vs sample rate.

Where does this definition of sample rate come from?

It rests on Wikipedia, Sampling (signal processing), read on 22 October 2026. What each source supports is listed above, next to its link.

Can you give an example of sample rate?

At 16,000 samples per second the highest frequency that can be represented is 16,000 / 2 = 8,000 Hz. At 8 kHz telephone rate it is 4,000 Hz. Duration check: a 44.1 kHz Ogg with 441,000 samples lasts 10.0 s; divided by 48,000 it would read 9.19 s, about 8 in 100 short.

Related

Start transcribing free

Send Autrans into your next meeting, use it in the room, or upload a recording. Your first 30 minutes are free, given once, and no credit card is needed to start.

Last checked 22 October 2026. How we measure accuracy. Spotted a mistake? Write to hello@autrans.online.