Puhekoodaus

Puhekoodaus refers to the Finnish term for "speech coding" or "voice coding." It encompasses the techniques and technologies used to represent and transmit speech digitally. The primary goal of speech coding is to compress the speech signal efficiently while maintaining acceptable perceptual quality. This is crucial for telecommunications systems, enabling more conversations to be carried over a given bandwidth, and for voice recording and storage applications.

There are various types of speech coders, broadly categorized into waveform coders and model-based coders. Waveform coders aim to reproduce the waveform of the speech signal as accurately as possible. Examples include Pulse Code Modulation (PCM), Differential Pulse Code Modulation (DPCM), and Adaptive Differential Pulse Code Modulation (ADPCM). These methods are generally simpler and offer good quality at higher bitrates.

Model-based coders, on the other hand, use a model of the human speech production system. They analyze the speech signal to extract parameters like pitch, formant frequencies, and excitation characteristics, and then transmit these parameters. The decoder synthesizes speech based on these parameters. Linear Predictive Coding (LPC) is a prominent example of a model-based coder. These coders can achieve very low bitrates, making them suitable for applications where bandwidth is severely limited, although they may sometimes introduce speech artifacts or an unnatural "synthetic" quality. Modern speech coding often employs advanced techniques, including those based on machine learning and neural networks, to achieve even better compression and quality tradeoffs.