Presentation on theme: "Audio. 音訊原理 聲音由聲波構成，由介質傳遞 Human Ears The human ear is able to detect frequencies in the range between 20 Hz and 20 kHz –Upper limit decreases with."— Presentation transcript:
Human Ears The human ear is able to detect frequencies in the range between 20 Hz and 20 kHz –Upper limit decreases with increasing age –e.g. iobasics/a/test_tones.htm
Human Ears Comparison
音訊儲存格式 一般音訊格式 : –WAV 檔格式 Microsoft 開發，用於 Windows 系統，副檔 名.wav –AU 檔格式 Sun 公司開發，用於 Unix 系統， 副檔名.au –AIFF 檔格式 APPLE 公司開發，用於 Mac 系統，副檔名.aif
Digitizing Sound Sampling –The selection of the sampling rate If limiting of hearing is 20 kHz, a minimum rate of 40 kHz is required by the Sampling Theorem. The sampling rate of audio CDs is 44.1 kHz kHz is commonly used for Internet kHz for speech DAT (digital audio tape): 48 kHz
Digitizing Sound Quantization –It ’ s usually quantization levels for CD audio 16 bits –Under-sampling a pure sine wave An analogue signal will be coarsely approximated by samples that jump between just a few quantized values –Dithering When a small amount of random noise is added to the analogue signal before sampling
Digitizing Sound Quantization
Digitizing Sound Dithering –Dither is an intentionally applied form of noise, used to randomize quantization error noisequantization error
Digitizing Sound Dithering –Concept of adding noise to reduce noise that doesn't make a lot of sense.
Audacity Recording –Click on the red Record button to begin recording. –Click on the blue Pause button to pause the recording. Press it again to continue. –Click on the yellow Stop button to cease recording. The cursor will return to its previous position, before the recording was started.
Digitizing Sound Recording and Importing Sound –Sampling rate and sampling size –If level of signal is too low, then resulting recording will be quiet. –If level is too high, clipping will occur. Fig –Gain control can be used to alter level. –Automatic gain control
Compression Use a bank of filters to split signal into bands of frequencies; 32 bands are commonly used. The average signal level in each band is calculated, and using these values and a psycho-acoustical model, a masking level for each band is computed.
Compression MPEG Audio 3 layers –Layer 1: 192 kbps for each channel –Layer 2: 128 kbps for each channel –Layer 3: 64 kbps for each channel MP3 = MPEG-1 Layer 3 compression rate = 10:1
Compression Block Diagram for MP3 –Input audio samples are fed into the encoder. –The mapping creates a filtered and subsampled representation of the input audio stream. –A psychoacoustic model creates a set of data to control the quantiser and coding. sampling_frequencyfrequency specified (kHz) '00'44.1 '01'48 '10'32 '11'reserved
Compression –The quantiser and coding block creates a set of coding symbols from the mapped input samples. –The block 'frame packing' assembles the actual bitstream from the output data of the other blocks, and adds other information (e.g. error correction) if necessary.
Compression Block Diagram for MP3 –SMR: signal-to-mask ratio
Compression Block Diagram for MP3
Compression Polyphase Filter
Compression SMR –signal-to-mask ratio
Compression Document/Paper –ISO/IEC –REALIZATION OF A PSYCHOACOUSTIC MODEL FOR MPEG 1 USING GAMMACHIRP WAVELET TRANSFORM MatLab –http://www.petitcolas.net/fabien/software/mp eg/
Reference 聯盟教材 /Audacity.ppt#263,6,Before Recording ml ml recording/DitherExplained.pdf