Microphones are Input Devices that convert acoustic pressure variations in the air into electrical signals that can be recorded, transmitted or processed by computers and other electronic systems. The microphone's transducer responds to changes in sound pressure and produces a corresponding electrical output, which is subsequently amplified and converted into digital samples when connected to a computer's digital audio system. Microphones are used for voice communication, recording, broadcasting, music production, speech recognition, video conferencing, gaming, accessibility and environmental measurement. Their fundamental purpose is therefore acoustic sensing rather than simply "recording voices".
Different microphone transducer technologies respond to sound in different ways. Dynamic microphones use electromagnetic induction and are generally robust and capable of handling high sound-pressure levels. Condenser microphones use a diaphragm and electrically charged capacitor structure and generally require an appropriate power source, such as phantom power in professional audio systems. Electret condenser microphones are a common compact implementation in consumer electronics. MEMS microphones use miniature semiconductor fabrication techniques and are widely integrated into smartphones, laptops, earbuds and other compact devices. Each technology involves trade-offs involving sensitivity, noise, size, frequency response and power requirements.
Microphone characteristics include sensitivity, frequency response, self-noise, maximum sound-pressure level, dynamic range and directionality. Sensitivity describes the electrical output produced for a given acoustic input, while frequency response describes how strongly different frequencies are captured. Directional patterns such as omnidirectional, cardioid, supercardioid and figure-eight describe how sensitivity varies with the direction of incoming sound. A microphone with a wide frequency response is not automatically better for every purpose because recording quality depends on the source, room acoustics and desired result. Directionality can be deliberately selected to reduce unwanted sounds or capture particular acoustic sources.
Computer microphones ultimately require analogue-to-digital conversion for digital recording or processing. Sampling rate determines how frequently the analogue waveform is measured, while bit depth determines the numerical resolution used to represent each sample. According to sampling theory, a system must sample at more than twice the highest frequency intended for faithful reconstruction, subject to appropriate filtering. Common digital audio sampling rates include 44.1 kHz and 48 kHz, while professional systems may use higher rates. Bit depth affects quantisation noise and theoretical dynamic range, although practical performance is also constrained by analogue electronics and environmental noise.
Microphones are integral to speech recognition and voice interfaces. The microphone captures acoustic speech, after which signal-processing and machine-learning systems can perform tasks such as noise reduction, voice activity detection, feature extraction and speech recognition. Microphone arrays can estimate the direction of sound and support beamforming, allowing systems to emphasise a speaker while reducing unwanted noise. Multiple microphones are particularly useful in laptops, smartphones and conferencing systems because spatial information can improve speech capture in reverberant or noisy environments.
Acoustic environment is often more important than microphone specifications alone. Room reflections, background noise, mechanical vibration, wind and microphone placement can substantially affect captured audio. A high-quality microphone placed poorly in a reverberant room can produce worse results than a more modest microphone used correctly in a controlled environment. Pop filters, shock mounts, wind protection and acoustic treatment can address particular problems. Digital processing can suppress noise and echo, but aggressive processing can introduce artefacts or remove desirable speech characteristics.
Microphones differ substantially according to application. Studio condenser microphones may prioritise detailed sound capture, broadcast microphones may emphasise controlled speech reproduction and background rejection, while MEMS microphones prioritise compactness and low power consumption. Headsets place microphones close to the speaker's mouth to improve speech-to-noise ratio, while conference-room arrays may use multiple microphones distributed across a table or device. Measurement microphones are designed around calibration and predictable acoustic response rather than conversational use. These distinctions show why there is no universally ideal microphone.
Privacy and security are significant because microphones can capture conversations and ambient sound. Operating systems commonly provide application permissions governing microphone access, and users or administrators may disable or restrict access where necessary. Hardware indicators or physical mute switches can provide additional assurance on some devices, although their implementation varies. Microphones should also be distinguished from speakers: a microphone converts sound into an electrical or digital representation, while a speaker performs the reverse general function by converting electrical signals into acoustic output. Their role in computing spans simple communication through sophisticated audio acquisition and machine-perception systems.
