DOC PREVIEW
UCSB ECE 160 - Lec14

This preview shows page 1-2-16-17-18-34-35 out of 35 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 35 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

ECE160Spring 2007Lecture 14MPEG Audio Compression1ECE160 / CMPS182MultimediaLecture 14: Spring 2007MPEG Audio CompressionECE160Spring 2007Lecture 14MPEG Audio Compression2Psychoacoustics• The range of human hearing is about20 Hz to about 20 kHz• The frequency range of the voice is typically onlyfrom about 500 Hz to 4 kHz• The dynamic range, the ratio of the maximumsound amplitude to the quietest sound thathumans can hear, is on the order of about 120 dBECE160Spring 2007Lecture 14MPEG Audio Compression3Equal-Loudness RelationsFletcher-Munson Curves• Equal loudness curves that display the relationship between perceivedloudness (“Phons", in dB) for a given stimulus sound volume (“SoundPressure Level", also in dB), as a function of frequency• The bottom curve shows whatlevel of pure tone stimulus isrequired to produce theperception of a 10 dB sound• All the curves are arrangedso that the perceived loudnesslevel gives the same loudnessas for that loudness level ofa pure tone at 1 kHzECE160Spring 2007Lecture 14MPEG Audio Compression4Threshold of Hearing• Threshold of human hearing, for pure tones: if a sound isabove the dB level shown then the sound is audible• Turning up a tone so that it equals or surpasses thecurve means that we can then distinguish the sound• An approximateformula existsfor this curve:ECE160Spring 2007Lecture 14MPEG Audio Compression5Frequency Masking• Lossy audio data compression methods, such asMPEG/Audio encoding, do not encode some soundswhich are masked anyway• The general situation in regard to masking is as follows:1. A lower tone can effectively mask (make us unable tohear) a higher tone2. The reverse is not true - a higher tone does not maska lower tone well3. The greater the power in the masking tone, the wideris its influence - the broader the range of frequencies itcan mask.4. As a consequence, if two tones are widely separatedin frequency then little masking occursECE160Spring 2007Lecture 14MPEG Audio Compression6Frequency Masking Curves• Frequency masking is studied by playing aparticular pure tone, say 1 kHz again, at a loudvolume, and determining how this tone affectsour ability to hear tones nearby in frequency– One would generate a 1 kHz masking tone, at a fixedsound level of 60 dB, and then raise the level of anearby tone, e.g., 1.1 kHz, until it is just audible• The threshold plots the audible level for a singlemasking tone (1 kHz) and a single sound level• The plot changes if other masking frequencies orsound levels are used.ECE160Spring 2007Lecture 14MPEG Audio Compression7Frequency Masking CurveECE160Spring 2007Lecture 14MPEG Audio Compression8Frequency Masking CurveECE160Spring 2007Lecture 14MPEG Audio Compression9Critical Bands• Critical bandwidth represents the ear'sresolving power for simultaneous tones orpartials– At the low-frequency end, a critical band is less than100 Hz wide, while for high frequencies the width canbe greater than 4 kHz• Experiments indicate that the critical bandwidth:– for masking frequencies < 500 Hz: remainsapproximately constant in width ( about 100 Hz)– for masking frequencies > 500 Hz: increasesapproximately linearly with frequencyECE160Spring 2007Lecture 14MPEG Audio Compression10Critical Bands and BandwidthECE160Spring 2007Lecture 14MPEG Audio Compression11Bark Unit• Bark unit is defined as the width of one criticalband, for any masking frequency• The idea of the Bark unit: every critical bandwidth is roughly equal in terms of BarksECE160Spring 2007Lecture 14MPEG Audio Compression12Temporal Masking• Phenomenon: any loud tone will cause thehearing receptors in the inner ear to becomesaturated and require time to recover• The louder is the test tone, the shorter it takesfor our hearing to get over hearing the masking.ECE160Spring 2007Lecture 14MPEG Audio Compression13Temporal and Frequency MaskingECE160Spring 2007Lecture 14MPEG Audio Compression14Temporal and Frequency Masking• For a masking tone that is played for a longer time,it takes longer before a test tone can be heard.Solid curve: masking tone played for 200 msec;Dashed curve: masking tone played for 100 msec.ECE160Spring 2007Lecture 14MPEG Audio Compression15MPEG Audio• MPEG audio compression takes advantage ofpsychoacoustic models, constructing a large multi-dimensional lookup table to transmit masked frequencycomponents using fewer bits• MPEG Audio Overview1. Applies a filter bank to the input to break it into itsfrequency components2. In parallel, a psychoacoustic model is applied to thedata for bit allocation block3. The number of bits allocated are used to quantize theinfo from the filter bank - providing the compressionECE160Spring 2007Lecture 14MPEG Audio Compression16MPEG Layers• MPEG audio offers three compatible layers :– Each succeeding layer able to understand the lowerlayers– Each succeeding layer offering more complexity in thepsychoacoustic model and better compression for agiven level of audio quality– Each succeeding layer, with increased compressioneffectiveness, accompanied by extra delay• The objective of MPEG layers: a good tradeoffbetween quality and bit-rateECE160Spring 2007Lecture 14MPEG Audio Compression17MPEG Layers• Layer 1 quality can be quite good - provided acomparatively high bit-rate is available– Digital Audio Tape typically uses Layer 1 at around192 kbps• Layer 2 has more complexity; was proposed foruse in Digital Audio Broadcasting• Layer 3 (MP3) is most complex,and was originally aimed at audio transmissionover ISDN lines• Most of the complexity increase is at theencoder, not the decoder - accounting for thepopularity of MP3 playersECE160Spring 2007Lecture 14MPEG Audio Compression18MPEG Audio StrategyMPEG approach to compression relies on:• Quantization• Human auditory system is not accurate within the widthof a critical band (perceived loudness and audibility of afrequency)MPEG encoder employs a bank of filters to:• Analyze the frequency (“spectral") components of theaudio signal by calculating a frequency transform of awindow of signal values• Decompose the signal into subbands by using a bank offilters(Layer 1 & 2: “quadrature-mirror"; Layer 3: adds a DCT; psychoacoustic model: Fourier transform)ECE160Spring 2007Lecture 14MPEG Audio Compression19MPEG Audio Strategy• Frequency masking: by using a psychoacousticmodel to estimate the just noticeable noise level:– Encoder balances the masking behavior and theavailable


View Full Document

UCSB ECE 160 - Lec14

Download Lec14
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lec14 and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lec14 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?