Comparing MP3, AAC, WAV, and FLAC: Digital Audio Formats Explained

Digital audio quality depends on two key parameters: sample rate and bit depth. Sample rate determines how many times per second audio is captured, measured in Hz (e.g., 44,100 Hz, 48,000 Hz, 96,000 Hz). Higher sample rates capture more detail but require more computer processing power. Bit depth determines the precision of audio data representation, measured in bits (e.g., 8-bit, 16-bit, 24-bit). Higher bit depth provides more detailed audio with less noise. The industry standard typically starts at 48,000 Hz sample rate with 16-bit or 24-bit depth.

Sample rate (measured in kHz) determines how many times per second an audio signal is sampled, with higher rates providing more accurate frequency reproduction; bit depth (measured in bits) determines the number of possible amplitude values per sample, with higher depths offering greater dynamic range and cleaner audio. For percussion recordings, recording at 24-bit depth and 48 kHz sample rate is recommended, as 44.1 kHz is standard for CDs while 48 kHz is standard for video audio production.

Digital audio is defined by sampling rate (measurements per second) and bit depth (precision per sample). Common rates are 44.1 kHz, 48 kHz, 96 kHz, and 192 kHz. Each additional bit provides ~6 dB dynamic range (16-bit = 96 dB, 24-bit = 144 dB). Higher sampling rates allow gentler anti-aliasing filters, reducing phase distortion. The Nyquist-Shannon theorem requires sampling at least twice the highest frequency (40 kHz for 20 kHz hearing range). Professional studios use 24-bit recording to maintain headroom for processing.

Digital audio recording relies on two fundamental specifications: sample rate and bit depth. Sample rate determines how many times per second the audio interface captures snapshots of the analog signal, analogous to a camera's frame rate. Bit depth determines the resolution or accuracy of each sample, representing how much information is captured. Higher bit depths capture more detailed information, creating more accurate representations of the original sound. Together, these specifications define the quality of digital audio recordings.

Data compression reduces file sizes to save storage and bandwidth. Two fundamental principles exist: (1) Lossless compression preserves all data for perfect reconstruction, using techniques like Huffman coding and LZW. (2) Lossy compression achieves higher ratios by removing imperceptible data, used in MP3 and JPEG formats. Compression is essential because storage and bandwidth resources are limited, making efficient data representation critical for handling large files like images, videos, and audio.

Data compression is a fundamental computer science concept that reduces file sizes for efficient storage and transmission. The video explains that compression saves disk space by reducing file sizes and saves bandwidth and transmission time by making data transfer faster. Using practical analogies like folding a blanket for storage and pushing a folded blanket for transmission, the instructor demonstrates how compression makes data more compact and manageable. The video emphasizes that compression is used daily for images, audio, video, text files, folders, and directories through software tools and online services.

Data compression reduces file sizes without losing essential information by exploiting patterns in data. The core principle is variable length coding, where frequently occurring elements receive shorter codes while rare elements receive longer codes. This concept originated in the 19th century with Morse code, where common letters had shorter signals. In digital systems, compression reduces the total number of bits (0s and 1s) needed to represent information. The fundamental insight is that data contains redundancy—repeated patterns that can be compressed without changing the meaning. This principle enables modern digital communication, allowing us to stream videos and send photos efficiently.

Data compression reduces file size for efficient storage and network transmission. Three main purposes: saving storage space, easier file transfer over networks, and consuming less bandwidth. Bandwidth is the path through which data travels, similar to a road. Two main techniques exist: (1) Lossy compression permanently reduces file size but sacrifices data quality, with maximum compression around 90%, applied to images, sound, and videos using perceptual compression. (2) Lossless compression retains original data exactly when decompressed, with maximum compression around 50%, applied to text and images. Perceptual compression removes data imperceptible to humans - pixels for images, frames for videos, audio bytes for sound. Run Length Encoding (RLE) is a lossless technique that encodes consecutive identical elements into count-value pairs. RLE has a critical limitation: when data lacks consecutive identical elements, file size can increase instead of decrease.

Data compression reduces file sizes by recoding information to require fewer bits, using either lossless algorithms (exact recovery) or lossy algorithms (reduced precision based on human sensory limitations for audio and images), with the goal of optimizing storage and transmission efficiency.

Humans detect sounds within specific amplitude and frequency ranges. The hearing threshold is 20 micropascals (0 dB), with maximum comfortable levels around 120-150 dB before damage occurs. Frequency sensitivity varies among individuals, with some having better hearing (golden ears) and others damaged from loud sound exposure. Psychoacoustics is the branch of science studying how the ear and brain process sound information, connecting physical sound properties with human interpretation. This field encompasses cochlea function, hair cells, basilar membrane, and the auditory system's development and pathology.

The commonly cited 20 Hz to 20 kHz hearing range is inaccurate. Under quiet conditions, conscious hearing narrows to 60 Hz to 16 kHz, and under whisper-quiet conditions, it further narrows to 500 Hz to 6 kHz. The 2-3 kHz region is where hearing is most acute. At extreme loud volumes (95 dB+), frequencies below 20 Hz become audible, but the required sound pressure levels would cause structural damage. Music reproduction systems typically cannot reproduce frequencies below 7 Hz with high quality, and there is minimal recorded material below this threshold.

Psychoacoustics is the science that studies the connection between perception and sound. Humans are very susceptible to certain frequencies - low frequencies always cause excitement, alarm, and distraction. Historically, this is related to the fact that before electronics, low frequencies meant threats like earthquakes, storms, or large rocks rolling toward you. The spectral center of human hearing is concentrated in the range of 2-4 kHz (1000 Hz), which is a very unpleasant sound that attracts attention maximally.

Humans can hear sounds within the 20Hz to 20kHz frequency range, where sounds below 20Hz are called infrasound and sounds above 20kHz are called ultrasound (used by bats/dolphins for echolocation and in medical imaging); age-related hearing loss typically begins at higher frequencies and progressively worsens over time, with approximately one-third of Americans aged 65-74 experiencing hearing loss and nearly half of those over 75 having significant difficulty hearing.

The human ear can perceive sounds from 20 Hz to 20,000 Hz (20 kHz), which is divided into eight distinct frequency bands: subgraves (20-40 Hz), graves baixos (40-80 Hz), graves altos (80-160 Hz), médios graves (160-500 Hz), médios (500-2,000 Hz), médios agudos (2,000-4,000 Hz), agudos (4,000-8,000 Hz), and super agudos (8,000-20,000 Hz). These can be grouped into three broader categories: low frequencies (sons graves), mid frequencies (sons médios), and high frequencies (sons agudos). Each frequency band contributes specific sonic characteristics—low frequencies provide body and presence, mid frequencies offer clarity, and high frequencies add brightness—but can also introduce unwanted qualities like muddiness or harshness if not properly balanced.