1/30/2016

SoundQualityvs.DataRate

TheScientistandEngineer'sGuideto
DigitalSignalProcessing
ByStevenW.Smith,Ph.D.
Home

TheBookbyChapters

AbouttheBook

BookSearch

StevenW.Smith

Blog

Contact

Chapter22AudioProcessing/SoundQualityvs.DataRate

Chapter22:AudioProcessing
Downloadthischapterin
PDFformat

SoundQualityvs.DataRate

Chapter22.pdf

Whendesigningadigitalaudiosystemtherearetwoquestionsthatneed
tobeasked:(1)howgooddoesitneedtosound?and(2)whatdatarate
canbetolerated?Theanswertothesequestionsusuallyresultsinoneof
threecategories.First,highfidelitymusic,wheresoundqualityisofthe

Tableofcontents
1:TheBreadthandDepthofDSP
2:Statistics,ProbabilityandNoise
3:ADCandDAC
4:DSPSoftware
5:LinearSystems
6:Convolution
7:PropertiesofConvolution
8:TheDiscreteFourierTransform
9:ApplicationsoftheDFT
10:FourierTransformProperties
11:FourierTransformPairs
12:TheFastFourierTransform
13:ContinuousSignalProcessing
14:IntroductiontoDigitalFilters
15:MovingAverageFilters
16:WindowedSincFilters
17:CustomFilters
18:FFTConvolution
19:RecursiveFilters
20:ChebyshevFilters

greatestimportance,andalmostanydataratewillbeacceptable.Second,
telephonecommunication,requiringnaturalsoundingspeechandalow
dataratetoreducethesystemcost.Third,compressedspeech, where
reducing the data rate is very important and some unnaturalness in the
sound quality can be tolerated. This includes military communication,
cellular telephones, and digitally stored speech for voice mail and
multimedia.

Google Drive For

Work
Get Unlimited Online Storage,
Docs, Spreadsheets &
Presentations.

Table 222 shows the tradeoff between sound quality and data rate for
these three categories. High fidelity music systems sample fast enough
(44.1kHz),andwithenoughprecision(16bits),thattheycancapturevirtuallyallofthesoundsthathumansarecapableofhearing.This
magnificentsoundqualitycomesatthepriceofahighdatarate,44.1kHz16bits=706kbits/sec.Thisispurebruteforce.
Whereasmusicrequiresabandwidthof20kHz,naturalsoundingspeechonlyrequiresabout3.2kHz.Eventhoughthefrequencyrange
has been reduced to only 16% (3.2 kHz out of 20 kHz), the signal still contains 80% of the original sound information (8 out of 10
octaves). Telecommunication systems typically operate with a sampling rate of about 8 kHz, allowing natural sounding speech, but
greatlyreducedmusicquality.Youareprobablyalreadyfamiliarwiththisdifferenceinsoundquality:FMradiostationsbroadcastwitha
bandwidth of almost 20 kHz, while AM radio stations are limited to about 3.2 kHz. Voices sound normal on the AM stations, but the
musicisweakandunsatisfying.
Voiceonlysystemsalsoreducetheprecisionfrom16bitsto12bitspersample,withlittlenoticeablechangeinthesoundquality.This
can be reduced to only 8 bits per sample if the quantization step size is made unequal. This is a widespread procedure called
companding,andwillbe

21:FilterComparison
22:AudioProcessing
HumanHearing
Timbre
SoundQualityvs.DataRate
HighFidelityAudio
Companding
SpeechSynthesisandRecognition
NonlinearAudioProcessing
23:ImageFormation&Display
24:LinearImageProcessing
25:SpecialImagingTechniques
26:NeuralNetworks(andmore!)
27:DataCompression
28:DigitalSignalProcessors
29:GettingStartedwithDSPs
30:ComplexNumbers
31:TheComplexFourierTransform
32:TheLaplaceTransform
33:ThezTransform
34:ExplainingBenford'sLaw

Howtoorderyourown
http://www.dspguide.com/ch22/3.htm

discussedlaterinthischapter.An8kHzsamplingrate,withanADCprecisionof8bitspersample,resultsinadatarateof64kbits/sec.

1/2

1/30/2016

hardcovercopy
Wouldn'tyouratherhaveabound
bookinsteadof640loosepages?
Yourlaserprinterwillthankyou!
OrderfromAmazon.com.

SoundQualityvs.DataRate
Thisisthebruteforcedataratefornaturalsoundingspeech.Noticethatspeechrequireslessthan10%ofthedatarateofhighfidelity
music.
Thedatarateof64kbits/secrepresentsthestraightforwardapplicationofsamplingandquantizationtheorytoaudiosignals.Techniques
forloweringthedataratefurtherarebasedoncompressingthedatastreambyremovingtheinherentredundanciesinspeechsignals.
Data compression is the topic of Chapter 27. One of the most efficient ways of compressing an audio signal is Linear Predictive
Coding(LPC), of which there are several variations and subgroups. Depending on the speech quality required, LPC can reduce the
dataratetoaslittleas26kbits/sec.WewillrevisitLPClaterinthischapterwithspeechsynthesis.
NextSection:HighFidelityAudio

Home | TheBookbyChapters | AbouttheBook | StevenW.Smith | Blog | Contact

Copyright19972011byCaliforniaTechnicalPublishing

http://www.dspguide.com/ch22/3.htm

2/2