SpeechViewer III Support Information

IBM SpeechViewerTM III for Windows® Support Information
Version 1
September 1, 1997
Specifications, Tested Systems, and Technical Notes

Special Needs Systems
11400 Burnet Road
IZIP 9448
Austin, TX 78758

Table of Contents

Overview of SpeechViewer III

System Requirements

IBM Laptop and Desktop Compatibility

Technical Support (Restrictions and Other Notes)

Overview of SpeechViewer III

SpeechViewer III is an IBM Independence SeriesTM product for speech/language pathologists, teachers, and other professionals to use in modifying speech patterns.

SpeechViewer III (SPV III) consists of:

It uses visual and auditory feedback to analyze and improve the speech skills of people who have speech, language, or hearing disorders. It is also useful for people who are trying to modify inflection, pronunciation, and vocal quality when learning new languages or improving an accent. Using SPV III, speech professionals and their clients can monitor and gain control over such speech attributes as voicing, pitch, loudness, phoneme accuracy, and speech timing.

SpeechViewer III is a sophisticated, clinical tool that takes advantage of the high-resolution graphics and high-performance sound capability in today's personal computers. It has a menu-bar format for access to speech exercises, point-and-click capability, and enhanced phoneme model creation.

Here are some features of the program:

System Requirements

To use SPV III, the following minimum machine requirements must be met:

Hardware:

In addition, your computer must have the following items installed:

Memory:

16 MB
Windows 95
12 MB
DOS & Windows 3.1
16 MB
OS/2® Warp Version 3

Hard file space:

8 MB
Minimum SPV III installation
32 MB
Full SPV III installation
60 MB
Full SPV III installation with space for approximately 30 client files

Software:

The appropriate audio and display device drivers must be installed.

The SoundBlaster driver must be at least version 3.04 or higher.

There is no specific level of Mwave driver required.

Drivers are now available to allow certain ESS multimedia-based systems to work with SpeechViewer III. (see reference #1 for specifications)

One of the following operating systems is required:

IBM Laptop and Desktop Compatibility

IBM ThinkPad®, Aptiva® and PC Desktop Compatibility

Many IBM ThinkPad products capable of supporting multimedia applications work well with SPV III. Within the ThinkPad product family, different brands of multimedia support are offered. The accompanying chart and support notes identify compatibility relating to ThinkPad and desk top models and associated software driver requirements. Some ThinkPads do not work with SPV III at all because of internal noise which bleeds into the SPV III audio and makes recording phoneme models and waveform analysis unusable. In addition, some ThinkPad and Aptiva models use Crystal Semiconductor multimedia chips/cards. SNS is working to achieve compatibility on these models, and will update this compatibility data as it becomes available.

IBM ThinkPads
A text equivalent of the IBM ThinkPad table is also available.
Model Requirements Comments
310
Model for Europe. Not tested.
360/365
Not recommended for use with SPV III. Internally generated noise interferes with phoneme model production at high frequencies.
380, 385
Not Compatible -Crystal Semiconductor Multimedial software drivers not compatible with SPV3.
701
Not recommended for use with SPV III. Internally generated noise interferes with phoneme model production at high frequencies.
560
  • Use CD-ROM drive a OEM option via PCMCIA or parallel port.
  • ESS minimum driver level*.
This machine is compatible with SPV III and can be used without the CD-ROM after full software installation.
750 Dock 1 or 2 station with sound card* and CD-ROM. This is the lowest recommended system due to 33 MhZ processor speed. Slow graphic display affects usability in a therapy setting.
755 C, Cs Dock 1 or 2 station with sound card* and CD-ROM. Processor speed is better than 750, but still negatively affects the graphic displays.
755 CE, CSE, CX with DSP, CV Dock 1 or 2 station with CD-ROM, or UltrabayTM or OEM option via PCMCIA or parallel port. Entry point systems for SPV III/reliable speech therapy. Can be used without dock station after full software installation.
755 CD, DCV 16MB DRAM Compatible with SPV III.
760 C, E, Dock 1 or 2 station with CD-ROM, or Ultrabay or OEM option via PCMCIA or parallel port. Compatible with SPV III. Can be used without the dock station after full software installation.
760 L, LD, EL ESS minimum driver level*. Compatible with SPV III.
760 ED, ELD
Compatible with SPV III.

IBM Desktop PCs
A text equivalent of the IBM Desktop PCs table is also available.
Model Requirements Comments
PC350-P90 Compatible Sound Card Tested Sound Blaster 16, AWE32, and Mwave all are compatible with SPV III.
IBM ValuePointTM 486/66Mhz Compatible Sound Card Tested Sound Blaster 16 and is compatible with SPV III.
PC-365, P-Pro 180 Compatible Sound Card Tested Sound Blaster 16 and is compatible with SPV III.
Aptiva M71-P133, A92-P166 Compatible Sound Card
Tested as configured with an IBM Mwave card and is compatible with SPV III.

Multimedia sound cards and on-board sound chips supported:

  1. Sound Blaster 16, minimum driver level is 3.04

  2. IBM Mwave, MDSP2780 DSP

  3. ESS 1688, 1868 with minimum driver level as follows:

During translation activities, SPV III was tested successfully on the following three machines:
A text equivalent of this Tested Machines table is also available.

Hardware


Make Aptiva Apricot Escom
Model 2155 MS 540 Full Tower
Memory 16Mb 16Mb 32Mb
Sound Card Creative Labs 32 AWE Media Vision Pro Audio Aztec Washington
CD ROM Mitec 4 speed Sony CDU76ES Mitsumi LU005S

Technical Support (Restrictions and Other Notes)

Update - 2/5/97

We have updated The US English Version of SPV III to the 1.02 level. Non US English versions level 1.0 are equivalent to the US 1.02 version.

The updates in level 1.02 are as follows:

  1. One highlight appears on the client setting screen instead of two.
  2. Shift+L has been added as a short cut key for the time scale in Create Phomeme Models.
  3. Eliminated a start record error message that appeared after the cheering reinforcement in Two Phoneme Contrasting (this occurred when running on systems faster than Pentium 90 MHz).
  4. Eliminated a problem with the course not moving forward in Two Phomeme Contrasting after the speed or reinforcement setting was changed.
  5. A Page Down (PgDn) function has been provided in Pitch & Loudness Patterning after saving a speech sample.
  6. On fast Pentium machines (Pentium 133MHz or greater), the audio and graphic reinforcement timings have been revised to maintain synchronization.
  7. Enabled all hypertext links within the help panels.

The updates that were included in level 1.01 are as follows:

  1. F6 is now listed and working as the accelerator for Compaction Threshold in the Combine Phoneme Model Files dialog box.
  2. The status line in Voice Timing reflects the appropriate action based on the current speaker setting (whether the speaker was changed using the keys or the menu system).
  3. The default recognition threshold was changed to 40 for the Phonology exercises.
  4. The new version number (1.01) appears in the About box.
  5. Voice Timing statistics (duration values) have been revised.
  6. Phonology exercises have been revised to handle large model files.
  7. Item 1 under Installation has been fixed.

Note that major SPV III restrictions and unique features are documented in the SPV III Guide as Notes within all of the sections and in Appendix A: Configuring Your Computer and Appendix B: Trouble Shooting. Please thoroughly review these appendices.

Below are some additional restrictions with details and explanations that may not appear in the Guide:

   Installation

  1. If you have SPV III version 1.0 and do not have an AUTOEXEC.BAT in WIN 95, installation fails. To remedy the problem, create an AUTOEXEC.BAT in the root directory of the operating system drive and then start the SPV III SETUP program. This problem was fixed in version 1.01. (Ref #1.)

Sound Card Device Drivers

  1. Creative Labs Sound Blaster Drivers:

    If the Sound Blaster 16 device driver is not at the 3.04 level, an update device driver will need to be installed to use with SpeechViewer III. To get the latest SB16 device drivers, use the following web site: http:\\www.creaf.com/wwwnew/tech/ftp/ftp-sb16awe.html. For WIN 3.1, the driver ZIP file should be s16dw3up.exe. For WIN 95, the driver ZIP file should be sbw95up.exe. You can also get SB16 drivers from the Creative Labs BBS at (408)428-6666.

  2. ESS Technology, Inc.:

    ESS drivers are available for IBM ThinkPad's from the IBM BBS at 919-517-0001 or the main IBMPC web site (www.us.pc.ibm.com/searchfiles.html) and they are in directory 23. On this page is a search tool and you can search on SpeechViewer, ESS, or the file name and it should return the file listing. The files are:

    All of these files are self-extracting executables and will ask for a blank formatted diskette.

  3. We have noted unusual behavior with drivers for the IBM 16-bit Sound Card under Windows 95. Follow these steps for a successful upgrade of the ESS drivers for this card:

    1. Install the sound card and drivers as described in the Options by IBM 16-bit Sound Card User's Guide.
    2. Get the latest ESS drivers from IBM Special Needs Systems.
    3. Use the Device Manager page of the Windows 95 System Settings to remove all sound drivers.
    4. Use the Add New Hardware feature from the Windows 95 Control Panel to add the updated ESS drivers.

      Note: Do not let Windows 95 search for new hardware. Force it to use the drivers you have on a diskette.

      At this point you should be able to use the sound card with the new drivers.

    5. During the next boot up, Windows 95 will say that it has found new hardware as it doesn't recognize the drivers that you installed. It will install another set of drivers. These will not be a problem even though they conflict with the correct drivers that you installed in the previous step. Simply leave them installed as they cause no problem because they are ignored. If you remove them, Windows 95 will just install them again during the next boot up.

  4. In general, we do not support Sound-Blaster compatible cards such as Reveal, Jazz 16, Vibra, and Diamond. Speech playback on these cards is choppy instead of smooth and continuous. We also do not support the M-ACPA card (from SpeechViewer II) or Digispeech.

    The current models of the IBM Aptiva desk-top and ThinkPad 380 series of computers use multimedia chips from Crystal Semiconductor. At the present time, the multimedia cards/chips in these machines are not compatible with SpeechViewer III.

  5. On a Win 3.1 system with Options by IBM Mwave sound card, you cannot use the Sound Recorder after using SPV III. You must restart Windows to use the Sound Recorder.
  6. On a system with an Mwave sound card, there may be a delay after you press a phoneme button in a phonology exercise before you hear the phoneme playback. This delay is not noticeable on SoundBlaster systems.

Display Device Driver

  1. PLEASE NOTE: SPV III is designed to be used at 640x480 resolution and 256 colors. In all cases where SNS has conducted compatibility tests on IBM systems, where available, there have been no display resolution compatibility issues when using the 640x480x256 resolution. Depending on display drivers, SPV III will work at higher resolutions. However, actual display image becomes smaller with increasing display resolution.
  2. If you change the system fonts, especially if you make them larger, the SPV III text will not fit properly on the controls. In addition, if the balloon always appears inflated in the Loudness Range exercise or if spectra do not appear in the Spectra Patterning exercise, your display driver is probably set up to support more than or less than 256 colors. You should run with system default fonts and the screen resolution set to 640x480 by 256 colors for the SPV III screens to display properly. Check the setting for the display device driver in the Windows Setup program or folder and select a display device driver that supports exactly 256 colors.
  3. If your system uses an ATI display driver, you may observe two problems:

    If you encounter this behavior, you must upgrade your display driver. You can download the latest drivers from HTTP://WWW.ATITECH.COM.

Palette Animation

  1. If you try to switch between SPV III and another graphics program, you may see strange colors temporarily. Switch back to SPV III, return to the exercises menu, and restart your exercise. The proper colors should be restored.
  2. In Win 95, if you bring up a help panel that has graphics (like a help panel for an exercise or the exercises menu help panel), you may see strange colors on the SPV III screens. When you leave help, the screen colors should be properly restored.
  3. You may occasionally see a white exercise button on the exercises menu instead of the proper exercise button. Return to the office screen and then back to the exercises menu and the icons should be restored.
  4. If the user changes the system colors to colors other than the default system colors, SPV III screen colors may look strange. Refer to the Appendix B: Troubleshooting in the SPV III Guide to solve this problem.

System Resources

  1. In the Voice Onset exercise, you cannot click on menu items or buttons while you are voicing into the microphone. Clicking on menu items or dragging the ending graphic may seem slow because SPV III does not allow Windows to handle these functions at all times. This is a trade-off between exercise speech input and graphics performance versus keyboard/mouse input response time.
  2. If you do not hear audio reinforcements when the Reinforcement setting is Graphics and Sound, then your system may not have enough available resources. See Appendix B: Troubleshooting in the Guide for ways to handle this problem.
  3. If you cannot get help for a screen, your system may not have enough available resources. However, there is no help available for save and get file dialogs and message boxes. See Appendix B: Troubleshooting in the Guide for ways to handle resource problems.
  4. If you are using SPV III in Windows 3.1 or WIN/OS2, and reinforcements are set to Graphics Only or Graphics and Sound, reinforcement is slightly different in the Pitch Control exercise than in other exercises. The graphic reinforcement is abbreviated and there is no sound for the baby, scuba diver, and car.

Microphone

  1. If you have to shout into the microphone to animate the graphics or you have any trouble with the Loudness Range exercise, then ensure your AGC setting (for SB16) in your mixer applet is checked and your microphone input is at the maximum level. Be sure you have a windscreen cover on your microphone and be sure you are using a supported sound card (many of the SB-compatible cards will not support SPV III). See Appendix A: Configuring Your Computer in the Guide.

    One mixer applet that does not appear in Appendix A is the Win 95 Creative Mixer for the Awe 64. In that mixer applet, to set AGC: Click start, programs, Sound Blaster AWE 64, Creative Mixer. When the mixer applet appears, right click Input/Output Settings and then click Microphone Gain Control.

    In addition, if your microphone appears to be too sensitive, be sure to use the windscreen. If it still seems to be too sensitive, you should check the mic gain in your mixer applet and adjust the gain (probably turn it down). You can check the sensitivity of your microphone using the SPV III Pitch & Loudness Patterning exercise. If the microphone is too sensitive, the waveform will be high and may be clipped at the top and bottom.

  2. Wind from your speech or from an external source like a fan can affect your pitch and loudness display (your speech sample shows a wide loudness range when you blow softly on your microphone), so be sure you have your windscreen on your microphone.
  3. If you purchase an SPV III upgrade for use on an IBM ThinkPad or other machine that has a sound card with a high-impedence microphone input jack:

Miscellaneous

  1. In Pitch Control, a white box appears if you vary your pitch too much. In SpeechViewer II, there was an X instead of a white box when you varied your pitch too much. See the help for Pitch Control.
  2. If you get RestartRecord errors but in general SPV III is working correctly, you may be clicking on menu items or buttons during audio reinforcements, during playback, or when you should be recording speech. During audio playback and when you see an hourglass, try to be patient and wait for the audio playback to complete.
  3. Occasionally you may see a pause in the animation at the beginning of a reinforcement or when you first start an exercise. Windows sometimes takes extra time to process user input and to redraw the screen.
  4. If you press Ctrl+Alt+Del without exiting and then try to restart, you could get a General Protection Fault. Always select File, Return to Windows to exit SPV III before you restart your system or Windows.
  5. Some of the Alt and shortcut keys for the menu items may not work from the Office or Exercises Menu. They do work after you use them once.
  6. Speech playback may seem clipped or out of order, especially in the Phoneme Accuracy or Multiphoneme Chaining exercises, because the audio reinforcement is resetting the internal speech buffer. If you would prefer to hear the complete playback rather than audio reinforcement, set Reinforcements to Graphics Only.
  7. In the File, Save As dialogs (for saving phoneme models, speech sample files, layout files, statistics, etc.), the existing filenames are greyed out. However, these existing files can be selected and replaced.
  8. On a WIN 95 system, if you configure your display with a title bar and menu bar smaller than 18 pixels each, the SPV III screen will have unwanted white space. You should reset your display to the Windows Default to display the SPV III screens correctly.
  9. In Create a Phoneme Model, you can move the cursors to segment the speech sample selected as the phoneme model. When you playback the phoneme model, you may hear a popping or abrupt start and end for the speech segment. To avoid this, you should record your phoneme (speech sample) within 500 milliseconds. Your speech sample cursors should be moved close to the start and end of your phoneme production, not somewhere in the middle.

Trademarks

IBM, SpeechViewer, Aptiva, OS/2, Mwave, Independence Series, Ultrabay, ValuePoint, and ThinkPad are trademarks or registered trademarks of International Business Machines Corporation. Click here for more information on IBM trademarks.

Microsoft and Windows are registered trademarks of Microsoft Corporation.

Other company, product, and service names may be trademarks or service marks of others.