This article explains how Krisp Microphone and Krisp Speaker work.
Krisp Microphone and Krisp Speaker are virtual devices. They are created on your computer when you install Krisp on it.
InfoA virtual device refers to a device file that has no associated hardware. It mimics a physical hardware device when, in fact, it exists only in software form. Therefore, it makes the system believe that particular hardware exists when it really does not.
Krisp Speaker keeps the human voice and removes all the other noises such as:
- animal and nature sounds,
- street, traffic and construction noise,
- keyboard and mouse clicks,
However, it detects and keeps the in-app sounds in the most popular applications and the telephony signals and voicemail beeps intact. Here are the most common telephony sounds that Krisp speaker doesn't cancel:
- Zoom - all the in-app sounds
- MS Teams - all the in-app sounds
- Google Meet - all the in-app sounds
- Cisco Webex - all the in-app sounds
- Skype - only default sounds
- Slack - only default sounds
Dial tones, busy tones, voicemail beeps are standardized for most of the commonly used applications. These sounds are also left intact by the Krisp Speaker.
Both Krisp Microphone and Krisp Speaker are one-channel devices. This means that the incoming and outgoing sounds will be mono.
InfoWhen you listen to a sound in stereo, each of your speakers plays a separate audio track. This gives you the impression of sound source localization. It means you can locate the position of a sound source within a space in 3D dimensions.
Mono playback systems use one speaker and can only produce a 2D perception of the sound localization. If two speakers are used, the same audio track can be played with both speakers mimicking the 3rd dimension.
If your voice goes through the Krisp Microphone, it will become mono, no matter how many channels your physical microphone has.
The other call participants will hear your voice in mono no matter if their physical speakers support stereo sound or not. If two speakers are used, your voice will go out both of them, but will still stay mono.
The quality of the voice is defined by the sampling rate, or by its frequency. When we talk about high-quality voice, it refers to voice transmitted at 32000hz or a higher sample rate. This is known as HD quality. In the meantime, the highest frequency audible for the human ear is 20000hz.
The sound can be HD no matter if it's mono or stereo.
Krisp Microphone supports HD voice in 32000hz sample rate if you have an appropriate physical device and if your computer allows that load. If the other call participants have high-quality speakers, they will hear your voice in HD. Learn more about the Krisp Noise Removal modes.
Krisp Speaker doesn't support HD voice. Regardless of the specs of your physical speaker, if you use it in Krisp, the highest sample rate for it is 16000hz. If you are not in professional sound production, this quality should not damage your experience.