Compression, recognition and synthesis of sound.
Speech and audio processing includes sampling, pre-emphasis, framing, MFCC features, linear prediction, codecs (MP3, AAC, Opus), noise reduction, beamforming, speech recognition and synthesis. Psychoacoustic masking enables lossy compression.
MP3 removes sounds the ear cannot hear because of masking.
Audio and speech processing turn sound into data for communication, recognition and synthesis, which shapes products from phones to assistants.
Compression, recognition and synthesis of sound.
Audio and speech processing turn sound into data for communication, recognition and synthesis, which shapes products from phones to assistants.
Compression that removes sounds the human ear cannot hear well, such as masked frequencies, to save bits.
Features are extracted from the audio and matched with acoustic and language models, often using neural networks in modern systems.
Typical applications include compressing music into a smaller file with a perceptual codec, recognising spoken commands on a smart speaker and synthesising speech for an automated announcement system.
Study "Image processing basics" first (topic 39 of 101 in Signals, Control & Comms). It covers the groundwork this topic builds on.
Short-time Fourier transform. Mel scale. Codec bit rate vs quality.
MP3 removes sounds the ear cannot hear because of masking.
The next topic is "Random signals and noise", which builds directly on this one.
It is classed as Intermediate in Signals, Control & Comms (topic 40 of 101). It is an intermediate topic. Work through the beginner topics in Signals, Control & Comms first.
Want to practise? Open the interactive course for quick checks, quizzes, simulations and progress tracking.