QE-10 User`s Manual
Contents
1. QuickEnhance LJ E 1 O OL OL 3 3 OdB INPUT SELECT o zn o ANALOG i 3 8 6 6 DIGITAL 12 S 12 INPUT LEVELS 48 18 DIGITAL INPUT oag 24 U 24 COAXIAL 36 36 R 48 48 TORE LEFT RIGHT Figure 2 QE 10 Input Section 3 1 INPUT SELECT SWITCHES There are three possible input sources to the QE 10 audio processor one analog and two digital See Sections 2 3 Digital Audio Interfaces and 2 4 Analog Audio Interface The INPUT SELECT and DIGITAL INPUT switches See Figure 2 are used to select which one of these three sources is routed through the QE 10 s audio processor With the INPUT SELECT switch in the ANALOG position the signal connected to the unbalanced RCA analog inputs is selected When this switch is in the DIGITAL position the DIGITAL INPUT switch determines whether the COAXIAL or TOSLINK digital input is selected This switch has no function when the INPUT SELECT switch is in the ANALOG position Note The INPUT SELECT switches only affect the routing of signals into the QE 10 All three outputs are always active and available regardless of the position of these switches In this way the QE 10 can function as an audio patchbay converting any of the three input types to any output 3 2 INPUT LEVELS The two INPUT LEVEL knobs See Figure 2 allow the user to add gain to a low level signal to bring it up to a normal processing level or to atten2. 17 W x 3 5 H x 8 D black aluminum case Optional rack mount kit available 30 front panel switches knobs and slider controls Stereo LED bar graph input level indicators 25 pin parallel printer interface connects to most ASCII capable printers for REPORT generation feature e 9 18VDC at 4A 2 1mm barrel connector positive tip External universal AC adapter included 16
3. 2 SETUP AND CONNECTIONS The QuickEnhance QE 10 has numerous ways to connect to common laboratory audio equipment including two digital audio interfaces and one analog audio interface This section details the type and purpose of the various rear panel connectors G QuickEnhance QE 10 uda Autre iua DIGITAL TAPE ENHANCEMENT SYSTEM ev DiBTAL Aub iG CORPORATION A Company SERIAL NuMBER QEI O il ANALOG ANALOG B PDIF TOSLINK OUTPUTS INPUTS DIGITAL INTERFACE DIGITAL INTERFACE DC POWER I TPUT INPUT S 18VDE 44 PRINTER PORT DUTPUT GQ ineur NN ZZ LL Borer ic Q err ze x sah Oee o umo g OO OO Figure 1 QE 10 Rear Panel 2 1 DC POWER A 2 1mm barrel type DC POWER connector See Figure 1 with positive tip is provided for supplying power to the unit An external universal AC adapter with line cord is included in the package The unit can accept DC power input at a range of 9 18V and draws a maximum peak current of 4A 2 2 PRINTER PORT The standard 25 pin parallel PRINTER PORT See Figure 1 is provided to implement the printed REPORT feature for forensic documentation of front panel settings See Section 5 2 Report Button The PRINTER PORT is suitable for connection to most inkjet dot matrix and laser printers that support ASCII text printing 2 3 DIGITAL AUDIO INTERFACES There are two types of DIGITAL audio interfaces provided on the QE 10 S PDIF on coaxial RCA conne
4. LES Once the input source has been selected and the correct input level set the signal passes on to the audio enhancement signal processing section of the QuickEnhance QE 10 The processing modules in the enhancement section provide the main functionality of the QE 10 Each of the modules is designed to remove specific types of noises and the modules can be used in various combinations to enhance audio with multiple noise problems The signal flows through the processing section of the QE 10 in the same order as the modules appear on the front panel of the unit reading left to right top to bottom HUM FILTER DECONVOLVER NOISE REDUCTION GRAPHIC EQUALIZER and AUTO GAIN The following sections describe these modules in detail explain the user interface controls for each and provide examples of common noise problems that may be improved by the use of the various modules MASTER OUT Figure 3 MASTER IN OUT Switch 4 1 MASTER SWITCH The MASTER IN OUT switch See Figure 3 is not a processing module per se but is used to enable or bypass the entire signal processing section of the QE 10 When the MASTER switch is in the IN position the QE 10 processes the selected input signal according to the settings of the various processing modules and passes the result to the audio outputs and headphone circuitry Conversely when the MASTER switch is in the OUT position audio from the selected input signal is passed directly t
5. QuickEnhance QE 10 Speech Enhancement Processor ASTER HUM FILTER DECGONVOLVER NOISE REDUCTION o e HEADPHONES REPORT LI vous ADAPT COMPLEX m mpm on eo our UT our on A nm MN MAK ADIIT ATTENUATION MONTOR wT LEFT 2 ce 1 6 D 0 o D u 4 E D 0 a 2 0 ma 100 o 2000 200 saca 4000 4800 a 8500 MD DIGITAL AUDIO CORPORATION USER S MANUAL V Digital Audio Corporation Technology to Make Listening Easier QuickEnhance QE 10 Speech Enhancement Processor USER S MANUAL Document Number 020610AME June 10 2002 L Digital Audio Corporation 4018 Patriot Drive Suite 300 Durham NC 27703 Phone 919 572 6767 Fax 919 572 6786 Web www dacaudio com Email sales dacaudio com Copyright 2002 by Digital Audio Corporation All rights reserved Table of Contents CHAPTER 1 INTRODUCTION eese eno ao eoe vo nono eo ane aeo nena apnea 4 CHAPTER 2 SETUP AND CONNECTIONS eere ee eren ee eee sees o etta a 5 21 IC ER 5 2 2 Printer Pott d 5 2 3 Digital Audio Interfaces sesssesssecssocesocesoosesocessccssocesocesoosssocesseessose 5 2 4 Analog Audio Interface essooesoossosesssccssecssocescocsoocesocessocesocesoosssossoso 6 CHAPTER 3 INPUT SECTION seccsssese sacsiessccestvcoessvncesecnsersonessseearesteesvbsnnesranees 7 3 1 In
6. UTO GAIN is already a good level no gain will be applied This module is useful for restoring the signal level of a heavily processed audio signal Also AUTO GAIN is invaluable for correcting the near party far party situation in which the talker closest to the microphone is picked up at a much higher level than the talker furthest from the microphone In fact any signal that has severely fluctuating audio levels can benefit from this module 13 CHAPTER 5 OUTPUT SECTION Once the audio signal is fully processed by the QuickEnhance QE 10 it will be sent to the audio outputs and headphone circuitry for listening or recording You may also wish to make a permanent record of the settings used to enhance a particular recording HEADPHONES REPORT VOLUME e C BUSY PRINT MONITOR POWER INPUT LEFT E E STEREO OUTPUT RIGHT 1 2 DIGITAL AUDIO CORPORATION Figure 9 QE 10 Output Section and Power Switch 5 4 HEADPHONES The HEADPHONES controls See Figure 9 on the front panel of the QE 10 give the user a convenient way to listen to audio signals passing through the device There are two 4 stereo phone jacks on the front panel that pass identical signals and are labeled 1 and 2 This is convenient for two person listening sessions or for training purposes The VOLUME knob controls the volume of signal in both HEADPHONES jacks equally Two MONITOR switches allow the user to select which signal is routed to t
7. ching the end of their useful life the recorder may have been running too slowly When the recording is played back on a tape player with fresh batteries or an AC power source the recording will appear to be running faster than normal causing the hum component to appear at a higher frequency than 50 or 60 Hz To handle situations like this the HUM FILTER base frequency can be tuned 10 via the ADJUST knob In other words for a base frequency of 50 Hz the HUM FILTER can be tuned in the range of 45 to 55 Hz and for a base frequency of 60 Hz a range of 54 to 66 Hz is possible To bypass the HUM FILTER stage so that it has no effect on the audio move the base frequency slider back to the OUT position DECONVOLVER CLEAR ADAPT m COMPLEX e ES FREEZE amp NORMAL OUT SIMPLE Figure 5 DECONVOLVER Module 10 43 DECONVOLVER MODULE The DECONVOLVER processing block See Figure 5 is designed to automatically adjust itself to remove noises that are time correlated from the input signal There are two main classes of time correlated noises that the DECONVOLVER can remove periodic repeating sounds such as tones hums engine idling sounds electrical motors etc and convolutional distortion such as echo reverberation and acoustic resonances caused by reflected sound in the environment the recording was made in There are three user controls in the DECONVOLVER module The CLEAR button is used to flus
8. ctors and optical digital on TOSLINK connectors See Figure 1 Both interfaces support input of stereo 24 bit digital audio signals at a maximum sampling rate of 96kHz These digital interfaces are commonly found on professional and some consumer audio equipment such as CD players MiniDisc players DAT machines PC sound cards etc Only one of the digital audio inputs can be active at a time See Section 3 1 Input Select Switches but both digital audio outputs are always active The digital outputs sampling rate is always 44 1kHz regardless of the sample rate of the digital input this is made possible by the high quality real time sample rate conversion in the QE 10 2 4 ANALOG AUDIO INTERFACE Unbalanced stereo ANALOG inputs and outputs See Figure 1 are provided on standard RCA connectors The ANALOG inputs are capable of accepting line level signals up to 3 0Vrms and have an input impedance of 25kO The ANALOG outputs have a maximum level of 3 0Vrms at full scale and the output impedance is 100 The ANALOG inputs are only active when selected with the front panel INPUT SELECT switches See Section 3 1 Input Select Switches but the outputs are always active CHAPTER 3 INPUT SECTION The input section of the QuickEnhance QE 10 is the place where the desired input source is selected and the audio input levels are set and monitored
9. h and re initialize the DECONVOLVER You may wish to do this whenever the signal you are processing has changed significantly and you want the DECONVOLVER to re acquire a fresh solution In very rare instances sudden jumps in the amplitude of the input signal may cause the DECONVOLVER module to become unstable resulting in static or harsh distortion at the output If this happens simply press the CLEAR button to correct this condition If the module continues to crash in this way it may be necessary to lower the input levels or to move the COMPLEX NORMAL SIMPLE slider to a different position to mitigate the problem This behavior does not represent a defect in the QE 10 but is an unavoidable side effect of the very powerful signal processing taking place inside the DECONVOLVER module The ADAPT FREEZE OUT switch is used to control whether the DECONVOLVER is actively adapting to the input signal freezing the current solution while ignoring changes in the input signal or not affecting the audio signal at all Depending on the type and nature of the noise problem you are trying to remove and the underlying signal of interest you may wish to allow the DECONVOLVER to continuously ADAPT through the entire recording or you may wish to train the module on a short portion of audio FREEZE the solution and apply that solution to the entire recording On difficult material you may need to try various approaches to see which technique yields the best
10. he HEADPHONES The INPUT OUTPUT switch routes either the currently selected INPUT source See Section 3 1 Input Select Switches or the final processed OUTPUT to the HEADPHONES The LEFT STEREO RIGHT switch routes only the LEFT channel signal to both ears the LEFT channel to the LEFT ear and the RIGHT channel to the RIGHT ear for STEREO listening or the RIGHT channel to both ears You may wish to use this control if you have a monaural signal plugged into only one input channel of the QE 10 Note The position of the MONITOR switches only affects the HEADPHONES outputs and has no effect whatsoever on the rear panel output signals Therefore the user may move these switches as desired without affecting any recording of the processed signal 14 5 2 REPORT BUTTON The REPORT button See Figure 9 is used in conjunction with the PRINTER port on the rear panel of the QE 10 See Section 2 2 Printer Port When a printer capable of producing ASCII text printouts is connected to the PRINTER port pressing the REPORT button will cause a plain text formatted REPORT containing the current status of all front panel processing settings to print There is a provision on the printed REPORT for the user to enter the name of the examiner the date and time the case number and other relevant information This REPORT can be used to fully restore a specific configuration of the QE 10 if needed 5 3 POWER SWITCH Use the POWER switch See Figure 9 to tu
11. hrough to the audio outputs and headphone circuitry bypassing all processing It is handy to use this switch to quickly compare the processed audio with the original unprocessed audio to determine the effect the QE 10 is having on speech intelligibility HUM FILTER o 50Hz 60Hz OUT 10 10 ADJUST Figure 4 HUM FILTER Module 4 2 HUM FILTER MODULE The HUM FILTER processing module See Figure 4 is useful for removing hum and buzz noises from the audio signal A typical source of hum noise is electrical AC power This hum may be caused by the proximity of the recorder microphone or microphone cables to AC powered electrical sources such as fluorescent lighting power cords motors home electrical wiring etc This noise is generally characterized by a strong energy component at either 50 Hz or 60 Hz depending on the power generation standard of the particular country the recording is made in along with decreasing energy components at integer multiples of the base frequency With the ADJUST knob at the center detent position move the HUM FILTER slide switch from the OUT position to either the 50 Hz or 60 Hz position to activate the HUM FILTER In doing so you may find that the base frequency of the hum does not fall at exactly 50 or 60 Hz resulting in incomplete cancellation of the hum For example if the recording was originally made on a battery powered analog tape recorder and the batteries were approa
12. n can be applied per band Move the IN OUT slide switch to the IN position to enable the GRAPHIC EQUALIZER module To reduce the high frequency components in the audio signal such as tape hiss the right most sliders may be moved down To reduce low frequency components such as low rumble the 12 left most sliders may be moved down Frequency bands that you do not wish to modify should be left at the top 0 dB position Note Moving the IN OUT slide switch to the IN position causes a fixed 300 Hz high pass filter to be inserted into the signal path This filter will remove any low frequencies which do not contain useful speech information and which may be limiting the intelligibility of speech signals in the recording AUTO GAIN IN 30dB ES 20dB OUT 10dB Figure 8 AUTOMATIC GAIN Module 4 6 AUTOMATIC GAIN MODULE The AUTOMATIC GAIN module See Figure 8 is a dynamic audio level processor which can be applied to the final output of the processing stage of the QE 10 The AUTO GAIN automatically increases the level of low amplitude audio signals as needed to maintain a good audio level at the output Move the IN OUT switch to the IN position to enable the AUTO GAIN module The AUTO GAIN module will apply gain as needed to maintain a good audio level up to a maximum amount specified by the Max Gain slide switch Three Max Gain selections are available 10 dB 20 dB and 30 dB If the signal input to the A
13. o enable this module The amount of attenuation the NOISE REDUCTION module applies to the random component in the audio signal is controlled by the ATTENUATION knob When the knob is rotated to the MIN position no noise attenuation is applied to the signal As the knob is rotated toward the MAX position more and more noise is removed from the audio signal Careful adjustment of this control is necessary to find the optimum position that causes the greatest amount of noise to be removed without removing any desired speech components se GRAPHIG EQUALIZER 8 s N f2 a e lt e T ee ee ses me 1G e e ees ses 24 re RR OUT 30 m a a a sos Kn rn 36 m _ a e oe os 1000 1500 2000 2600 3300 4000 4800 5650 6600 Hz Figure 7 GRAPHIC EQUALIZER Module 4 5 GRAPHIC EQUALIZER MODULE The GRAPHIC EQUALIZER module See Figure 7 is used to modify the tonal balance of the audio signal The signal is divided in frequency into ten bands spread roughly evenly through the portion of the frequency spectrum that human speech occupies Ten sliders one per band are used to attenuate signal energy in each band independently of the others Slider bars are positioned at 500 1000 1500 2000 2600 3300 4000 4800 5650 and 6500 Hz The lower a particular slider bar is moved the more attenuation is applied Up to 36 dB of attenuatio
14. put Select Switches eesssessecssocesocesoocesoecssecesocesooessocesseessocesocssooseoo 7 3 2 Input Levels aoi uito ae uoo be EIER c oe iain Fe vd oo ERE USMa MEER RR 8 CHAPTER 4 AUDIO ENHANCEMENT MODULES eee eere en 9 4 L Master Switch eot uas epiac n vk tuae ee Ph dere e exa ey eode ete FERRE TE eUR 9 4 2 Hum Filter Module eine hri onini ro kg SE eeu Susa Do a saab eo rk eek orba ce xod 10 4 3 Deconvol ver Module 5 25 5 5 i aapi ee Lise vo ded h Cae supo Tn SF Fe PE aC E TROU EE Te pep ona Dn 11 4 4 Noise Reduction Module ee eeee eere eere eee ee eene ee eee na aeo 12 4 5 Graphic Equalizer Module ee ee ecce eee eerte ee eere e eene ena 12 4 6 Automatic Gain Module eee eee ee eese ee eee eee ee ette seen seen uu 13 CHAPTER 5 OUTPUT SECTION ies eee veo reo n erepto Fe enu ero EN en Rn UR EH eva U ann 14 5 4 Headplhionbs u coii eickfe oia Ca depre beue iU kpP eR aal PR MERE Ide osoite REA ME dassi rias 14 5 2 Report Button 2 529 99 bist cedp Hai esie csi ede sce D da XE Eb DEM aga 15 5 5 Power Switch eet Pee tee eee REM EEN EI pisek oeart eioi EH EMEN NIRE e Ge to 15 CHAPTER 6 SPECIFICATIONS ieeesitinteecethe e nee ke cR CEU ee irte e a n Re INE Sere 16 Table of Figures and Tables Figure 1 OQE I0 Rear Panel isssisisscsscssisecssseassossecnsoviessencasasiconns evasborcasensessvncsines 5 Figure 2 QE 10 Input Section 5 ess
15. result The COMPLEX NORMAL SIMPLE mode switch allows the user to modify various parameters in the DECONVOLVER algorithm to better adapt to and remove the current noise problem The table below gives examples of the types of noise problems that are helped by each of the three settings With this in mind however you should always experiment with different settings to determine the optimum position for your particular enhancement situation Mode Example Usage Simple Removal of simple periodic noises such as tones and for the cancellation of minor echo reverberation and resonances in small rooms Normal Removal of more complex periodic noises such as buzzes hums idling engines etc and for the cancellation of echo reverberation and resonances in recordings made in small to medium sized rooms Complex Removal of significant echo reverberation and resonances in recordings made in medium to large sized rooms Table 1 DECONVOLVER Mode Selections 11 NOISE REDUCTION H IN OUT MIN MAX ATTENUATION Figure 6 NOISE REDUCTION Module 4 4 NOISE REDUCTION MODULE The NOISE REDUCTION module See Figure 6 unlike the DECONVOLVER is useful for removing noises that are not time correlated but are more random and unpredictable in nature Examples of this type of noise are air conditioner sounds wind noises radio static tape hiss etc Move the NOISE REDUCTION slide switch to the IN position t
16. rn the QE 10 on and off 15 ANALOG Line Inputs Line Outputs Headphones Output Analog Conversion DIGITAL DSP Processors Digital Inputs Outputs Audio Processing Processing Bandwidth CONSTRUCTION Enclosure User Controls Ports Power CHAPTER 6 SPECIFICATIONS Stereo e Panel adjustable 0 1 Vrms to 3 0Vrms Zin 25KQ Ground isolated RCA input jacks Stereo 3 0Vrms at full scale Zout 100Q Unbalanced RCA output jacks e Dual 4 stereo headphone jacks Selectable INPUT or OUTPUT monitoring Selectable LEFT channel RIGHT channel or STEREO monitoring VOLUME control 24 bit 128x oversampling stereo A D and D A 290dB SINAD Sigma delta conversion technology TMS320F206 20MIPS controller CPU 2 ADSP 2184 at 32 MIPS 4 DSP56200 digital filter processors Optical TOSLINK professional format digital input and output Coaxial S PDIF professional format digital input and output Supports sample rates of 32 44 1 48 and 96 kHz via asynchronous sample rate conversion Fixed 44 1 kHz sample rate on digital outputs Adjustable stereo 50 60 Hz Hum Filter with 1000 Hz notch limit Stereo 1024 tap 1CH Deconvolver with selectable presets Stereo automatic Noise Reduction filter with adjustable attenuation Stereo 10 band Graphic Equalizer with integral 300 Hz highpass filter Stereo Automatic Gain dynamic level processor 7 0 kHz fixed 16 4 kHz sampling rate
17. sssevcsdecacecsyecsebaadsdccescsyudessedsccevensvesenesdseacdesecsens 7 Figure 3 MASTER IN OUT Switch eee eese e eere ee eee eee eene enata etta ano 9 Figure 4 HUM FILTER Module 4 eeeee eese eee ee eee eee essen aeta setas eoe neco 10 Figure 5 DECONVOLVER Module eee eee ee eee eene eee eene en seen etta sensns 10 Figure 6 NOISE REDUCTION Module eere ee esee reete eene enne tn en 12 Figure 7 GRAPHIC EQUALIZER Module ee ceee eese rennen e etnee 12 Figure 8 AUTOMATIC GAIN Module A eeeeeee eene eene eee seen ena eo se tne 13 Figure 9 QE 10 Output Section 1 eere eerte ee eren ee ette eese ee seta ee eno 14 Table 1 DECONVOLVER Mode Selections c eeeeeee eee eee eene 11 CHAPTER 1 INTRODUCTION Thank you for purchasing the QuickEnhance QE 10 Speech Enhancement Processor We hope you will find this product the breakthrough in power ease of use and value that we intended it to be Please take a few moments to read this User s Guide to familiarize yourself with the main features and functions of the QE 10 If you are new to audio enhancement pay particular attention to Chapter 4 Audio Enhancement Processors which details the operation and recommended usage of the vast processing resources the QE 10 has to offer Enjoy using your new QE 10 CHAPTER
18. uate a loud signal that may be overloading the input circuitry of the QE 10 The center detent position of the knobs indicates the 0 dB point where no gain or attenuation is applied to the input signal These INPUT LEVEL knobs work the same regardless of the type of input selected See Section 3 1 Input Select Switches The input signal strength can be monitored using the LED bargraph display See Figure 2 This display is calibrated in units of decibels dB below full scale In other words a signal having peaks that illuminate the 18 segment of the bargraph display is 18 decibels below the maximum possible signal level into the QE 10 A normal processing level will cause the peaks of the input signal to illuminate the yellow segments of the display without illuminating the red overload OL segments If the input signal is too strong causing the OL indicators to illuminate rotate the INPUT LEVEL knobs to the left to attenuate the input signal If the input signal is too weak causing none of the yellow segments to illuminate rotate the INPUT LEVEL knobs to the right slowly to add gain to the input signal Note Be wary of sudden loud sounds in the input signal such as door slams gun shots etc which may cause the input circuitry of the QE 10 to overload This condition is highly unpleasant to listen to and may cause hearing damage if the listening volume is set too high See Section 4 1 Headphones CHAPTER 4 AUDIO ENHANCEMENT MODU
Download Pdf Manuals
Related Search