PCM (Pulse Code Modulation) is a method that processes an analog sound wave through sampling, quantization, and encoding, in that order, turning it into a sequence of 0s and 1s.
Analog sound wave
Sampling
Quantization
Encoding
Write the rounded values as 0s and 1s, producing digital data.
4 / 4
Sampling reads the height of the wave at fixed time intervals, quantization rounds each reading to one of a fixed set of levels, and encoding writes those numbers as 0s and 1s. Raising the number of samples per second (sampling frequency, 標本化周波数) and the number of bits used to express the levels (quantization bit depth, 量子化ビット数) brings the result closer to the original wave, but also increases the data size. The data size equals sampling frequency × bit depth × channels × seconds, so recording 10 seconds at 8,000 samples per second, 8 bits, and 1 channel gives 640,000 bits (80,000 bytes).
AnalogyIt is like taking burst photos of a runner at fixed intervals, reading each position off the nearest mark on a scale, and writing it down as a number.
How to rememberRemember the order as three actions: measure → round → write. The C in the name (Code) matches the last step, encoding.
Exam points
Mix-upSampling (標本化) divides the time axis, while quantization (量子化) divides the range of values, so each one makes a different dimension finer.
MIDI (Musical Instrument Digital Interface) is a standard for passing performance information (演奏情報) between electronic instruments and computers: not the sound waveform, but which note to play, when, and how strongly.
Because it sends only instructions and no sound itself, its data size is very small for the length of a piece. The actual sound is produced by the receiver's sound source (a device or software that generates sound) following the performance information, so the same data sounds different on a different sound source. A waveform itself, such as a recorded human voice, cannot be expressed as MIDI performance information.
AnalogyIt is like handing over sheet music rather than a recording: the same score sounds different depending on who plays it and on what instrument.
How to rememberThe last word of its full name is Interface. It began as an agreement for connecting one instrument to another.
Exam points
ExamMIDI carries instructions such as pitch, length, and loudness, and the receiving electronic instrument or computer produces the sound from them.
WAV is the standard audio file format on Windows, mainly used to store PCM-digitized audio without compression.
Because it is not compressed, no information is discarded, so saving does not change the sound quality, which makes it suitable as material for editing. The trade-off is that the PCM data size becomes almost exactly the file size.
AnalogyIt is like saving a photo exactly as taken, with no resizing or editing.
How to rememberThe name comes from Waveform. Picture a box that holds the waveform just as it is.
Exam points
Mix-upPCM is the method that turns analog audio into numbers, and WAV is the file format that holds those numbers: one is the method, the other the container.
MP3 is a lossy compression audio file format that makes audio data smaller by removing components the human ear has trouble hearing.
Lossy compression is a method in which discarded information does not come back even when the compressed data is decompressed (伸長, converted back to its original form). MP3 cuts parts that are hard to make out, such as a quiet sound right next to a loud one, and depending on the settings can shrink the data to about one tenth of the original PCM data. The more it cuts, the smaller the file, but the lower the sound quality.
AnalogyIt is like lightening a bag by taking out things you can do without when packing. What you take out cannot be recovered later.
How to rememberMP3 stands for MPEG-1 Audio Layer-3, a name taken from the audio part of MPEG, the compression standard for video and audio.
Exam points
ExamMP3 handles audio, and it makes files smaller through lossy compression (非可逆圧縮), which discards components that are hard to hear.
Mix-upData compressed with lossy compression cannot be fully restored even after decompression, while lossless compression (可逆圧縮) such as ZIP restores it completely.
AAC is a lossy compression audio format developed after MP3 that tends to keep better sound quality than MP3 at the same data size.
Audio file formats
Uncompressed
WAV
Lossy compression
MP3
AAC
Its name stands for Advanced Audio Coding, and it was defined as part of the MPEG-2 and MPEG-4 standards. It is used in music streaming services, smartphones, and the audio of digital broadcasting.
AnalogyIt is like an improved packing tool: at the same weight, it lets you keep more of the contents.
Exam points
Mix-upAAC and MP3 are both lossy compression formats; the main practical difference lies in how much sound quality each keeps at the same data size, and AAC usually has the edge.
Key points
PCM turns a sound wave into numbers through sampling, quantization, and encoding, in that order; higher sampling frequency and quantization bit depth bring it closer to the original sound but increase the data size.
MIDI data contains performance instructions, and the sound you hear is generated by the receiver's sound source.
WAV stores PCM data without compression, so there is no loss in quality, but files are large.
Information discarded by lossy compression does not return after decompression, and MP3 and AAC both use this method to make audio smaller.