MARCH 31, 2008.  Lawyers, judges, law students, legal administrators, and IS personnel visited the company booth at the ABA TECHSHOW  on March 13-14 at the Hilton Chicago.  Various uses for The Talking Form™ features of  SpeechMax™ HTML session file editor were demonstrated.  These included form creation and completion using speech recognition, traditional dictation, keyboard, or bar code.  Various legal-related uses of the software were discussed, including correction of speech recognition text by legal secretary or other assistant with a USB footpedal for audio playback.  Other demonstrations covered document assembly, case management, law review article and comment preparation, recording witness interviews, collection of medical and other expert opinion, audio mining, speech redaction, integration with Outlook, Word, and WordPerfect , multidocument text comparison, translation synchronization, and use with knowledge management systems.  It was also noted that revision history metadata was not stored in the XML session file.  General practice, solo, and small firm lawyers expressed particular interest in the software.    

APRIL 22, 2007.  The Medical Transcription Industry Association (MTIA) held its 18th annual conference on April 19-21, 2007 in Jacksonville, Florida.  Lee Stephen, programmer, was a featured speaker at the Vendor Showcase, and spoke on "Training the Transcriptionist of the Future."  He described techniques for training transcriptionists to use server-based speech recognition, and multiple ways in which SpeechMax™ can be used to create customized transcriptionist training files, including use of "speech censor" feature to delete confidential audio and text PHI (patient health information) to create sanitized training material.  If  would like further information on company products, review the pre-meeting eBlast, or email lstephen@customspeechusa.com.

MARCH 30, 2007.  "Dragon NaturallySpeaking Preferred 9 gives small business users and PC enthusiasts the power to create documents, reports, e-mails and more—all by speaking! Fast, easy and amazingly accurate, it’s over three times faster than typing. Just use your voice to dictate and edit in Microsoft® Word and Excel®, Corel® WordPerfect®, and virtually all Windows®-based applications. You can even dictate into a handheld device when you’re away from your PC, or use a Bluetooth microphone! It’s never been easier to use—no script reading required, so you can get started right away!"  Nuance website 03.30.07  Use the Preferred 9 version of the world's most popular software with SpeechProfessional™. 

MARCH 27, 2007. Full version of SpeechProfessional™ includes workflow management with web services and web reports for server-based speech recognition and general file management and document retrieval.   At $169, "Lite" version is intended for office that does not need the workflow features available in the full version.  The "Lite version" may be used with many Olympus, Sony, Philips, and other digital dictation and transcription products, Dragon, IBM, and Windows Vista speech recognition, and includes sound recorder, transcription audio playback, audio conversion, and speech-oriented text editor for speech recognition, text to speech, and other advanced speech processing.  This version supports playback of speech recognition files with standard, off-the-shelf USB transcriptionist footpedal.  Twin-speaker stereo Spectra headset with stereo plug for computer use and durable, three-position Infinity USB footpedal are available as combo package with "Lite" version for only $229.  

MARCH 19, 2007.  Orthopaedic Specialists of NW Indiana provides general orthopedic care and related specialty services. The group wanted to upgrade from a tape cassette dictation and transcription system to digital dictation with a handheld recorder.  A local physical therapy service had previously implemented a similar system.  The orthopedics group purchased PlayBax™ transcription playback software after reviewing use of the software by the local physical therapy service.

MARCH 17, 2007.  Vee Technologies provides business software outsourcing for insurance and medical claims processing and other activities, and recently purchased acWAVE™ SDK. Company clients are in insurance, health care, banking, logistics, and knowledge processing outsourcing.

MARCH 16, 2007.  M3 Medical Management Services provides per diem and travel nurse staffing, plus medical transcription, electronic medical record, billing and collection, and a nurse-operated answering service. Advisory council and executive team includes professionals with background in medicine, nursing, physiology, business, programming, and law. The company provides transcription services to many physicians at Cook County Hospital, many representing foreign medical graduates. As many of physician speakers are foreign born, the company recently purchased SpeechProfessional™ and SweetSpeech™ for server-based speech recognition to create custom speech user profiles that reflect the speaker's accent and speaking habits. Company plans to use SpeechMax™  with footpedal for editing the server-based speech recognition text.

MARCH 5, 2007.  Speech Strategy News March 2007 newsletter reviewed Custom Speech  development options.  The article referred to the "particularly wide range of applications" and recent update of software to run with Microsoft Vista speech recognition.  For example, the article noted, "With Custom Speech's SpeechMax loaded onto a system running the Vista operating system, a doctor, lawyer, or other speaker can dictate and view the transcribed text real-time . . . . Alternatively, the speaker can record dictation, and send it for transcription by Custom Speech's SpeechServers using Vista speech recognition."  For the entire review, click here.   

FEBRUARY 23, 2007.  ReadingWare, LLC is a 2006 recipient of a Small Business Innovation
Research grant from the National Education Center for Research of the Institute of Education Sciences, U.S. Department of Education. The firm specializes in creation of reading materials for preschool and kindergarten children to read on line, and for the dyslexic and blind. In 2006, it exhibited at the Los Angeles Times Festival of Books and the Miami Book Fair International. The company recently purchased SweetSpeech™ SpeechSplitter™ and SpeechMax™ to assist with preparation of segmented speech and text data for its audio books.

FEBRUARY 20, 2007.  DialAmerica Marketing, Inc., Mahwah, NJ. is the largest, privately-owned
telemarketing company in the United States with over 40 years of experience in the magazine industry and annual revenues of $185 million.   Seeking a solution for efficient audio conversion, the company recently purchased licenses for acWAVE™ standalone and batch processor.

FEBRUARY 19, 2007.   Billings Clinic (formerly Deaconess Billings Clinic) is the largest employer in Billings, Montana.  It includes a multispecialty group practice with over 200 physicians and nonphysician providers, with a structure similar to that of the Mayo Clinic.  Over the last several months the clinic extensively evaluated SpeechServers™ for Dragon onsite as a back-end add-on to their preexisting digital dictation and transcription system.  The server-generated speech recognition text was edited using the SpeechMax™ HTML text editor and speech processor.  The transcriptionist could navigate using a standard, three-position transcriptionist foot control.  After several months of testing, the clinic has issued purchased orders for additional copies of the software.  The sale was generated by long-time Market Partner Western Office Equipment.  Custom Speech USA provided programming scripts to assist with integration with the preexisting system and other technical assistance.

FEBRUARY 6, 2007.     CallDictate™ telephone dictation is available as a software or PC solution.  A 4-port Dialogic (Intel) telephony card is standard.  A large New York City law firm recently placed an order for a preconfigured CallStation™ PC system with a 16-port card.  Many of its lawyers dictate from cell phones.  Among its features, the system supports easy reentry into the dictation audio if the cellular network drops the call.  As further stated by a local digital dictation systems solution provider, "After reviewing several telephone dictation systems, we concluded that [CallDictate™] provided considerably more flexibility and opportunity or integration than competing applications. For our purposes, we felt that it as an ideal product to work as a back-end, add-on telephony solution for our digital dictation and transcription solution."

JANUARY 20, 2007.  Columbia University in New York City includes undergraduate, graduate, and professional schools, a school of continuing education, and a medical center.  Recently the medical center's biomedical informatics program began a pilot project to convert physician telephone queries for medical literature and other information into text using server-based speech recognition.  After discussion with the senior programmer analyst for the pilot, the informatics department purchased SpeechProfessional™.  This included SpeechServers™ for Dragon and related workflow software for a back-end, server-based add-on to a medical query system.   Cost and ease of integration with other applications were primary considerations in selecting the Custom Speech USA software.

DECEMBER 19, 2006.  TMA Associates publishes Speech Strategy News, an industry newsletter on the business, products, markets, and companies in speech recognition, text-to-speech synthesis, speaker authentication, and voice search.  The December issue contained a review of Custom Speech USA products and referenced the update of SpeechMax™ and SpeechServers™ to be compatible with Dragon 9.x. 

NOVEMBER 3-4, 2006.  The American Association for Medical Transcription (AAMT) presented its first ever Technology Vendor Showcase. Together with the Illinois Association for Medical Transcription (IAMT), the Missouri Association for Medical Transcription (MAMT), and the Business Issues Group, the organization provided a platform for emerging technology and educational advances in the healthcare industry. The event was held at the Sheraton Westport in St. Louis, Missouri on November 3-4, 2006.  Lee Stephen, Custom Speech USA programmer,  spoke on “Editing Real-Time and Server-Based Speech Recognition, and More.”  The talk explored editing speech recognition SpeechServers™ session files with the HTML text editor SpeechMax™. Both applications are included in the software suite SpeechProfessional™, which also includes workflow management, sound recorder, audio conversion, text to speech, transcription playback, and macro recorder.  Among other features, AAMT members found the text editor's SpeechCensor™ feature potentially useful as a way to create transcriptionist training material where the patient name, for example, has been deleted both from the text and the audio.  Using the patent-pending feature, the operator may select text and replace it with other text, for example "PATIENT NAME," and replace corresponding audio with a beep.

NOVEMBER 2, 2006.  Custom Speech USA presented its SpeechMax™ text editor solution at the Northwestern University's Medical Informatics program at a class held in Chicago.  The presentation was designed to compare graphical user interface for a speech-drive program compared to interfaces for electronic health records that typically use keyboard or mouse for data entry.  The forms creation and text compare features of the text editor were highlighted.  The class was attended by physicians, programmers, health care administrators, and other health informatics students.

OCTOBER 16, 2006.  For the Record (Great Valley Publishing Co., Inc.) is directed to health care administrators and other health users.  An article by Robbi Hess, "Physician Persuasion:  How to Sell Speech Recognition," deals with the issue of convincing physicians to use speech recognition.  It notes that front-end speech recognition potentially can reduce the time a physician spends entering notes in a medical record and provide near-zero turnaround time for dictation.  However, many doctors are reluctant to use the technology because they must correct the inevitable mistakes made the PC.  For this reason, back-end, server-based speech recognition has increased in popularity as a way to gain acceptance of this new technology in the medical community.  As explained in the article:

      Back-end speech recognition is the process by which the clinician dictates into a digital
      dictation system, the voice is routed through a speech-recognition machine, and the draft
      document is routed— along with the original voice file—to a medical transcription editor who
      verifies the accuracy of the draft, finalizes the report, and forwards it for signature.

      Lee Stephen, a programmer at Custom Speech USA, says back-end speech recognition
      should be easy for any physician with dictation skills. “The only difference is that the audio file
      is now transcribed first by a machine, not by a person,” he explains. “Front-end SR [speech
      recognition] is more difficult for some physicians [whose] word error rate is high and many
      corrections are required. For other physicians, where there is low error rate, there should be
      less resistance to adoption of speech recognition. However, it should be remembered that
      even a 5% error rate means an error about once every two sentences. If there is a short report
      with a dozen sentences, that still means that the physician may have to spend several
      minutes correcting the errors. For a busy physician committed to patient care, that is still a
      distraction.”

      Stephen believes physicians are most concerned that editing speech recognition errors will
      take time away from dealing with patients and other medical issues. “Consequently, a server-
      based system—where the speech recognition prepares a ‘rough draft’ for editing—
      and correction by a transcriptionist makes a lot of sense,” he says.

      Front-end speech recognition, Stephen says, is useful for completion of short reports in STAT
      settings where there is minimal correction required by the physician and/or there is a need for
      rapid turnaround time.

      “Where the report is longer or required turnaround time is longer, there is less need for front-
      end speech recognition,” he says. “Back-end speech recognition can be used, relying upon
      the transcriptionist editor to create a polished draft after correcting misrecognitions.”

Other interviewed were Nuance Communications, Inc. senior adviser Don Fallati and eScription Marketing Director Lauren Richman.

JUNE 13, 2006.  Updated version of SpeechMax™ is a multilingual, multiwindow HTML text editor that supports real-time, interactive speech recognition and server-based, back-end speech recognition with Dragon, IBM, Microsoft, and SAPI 5.x speech recognition engines. 
PRESS RELEASE

FEBRUARY 25, 2006.  Northern Illinois Physicians for Connectivity hosted an Information Technology Summit on February 25, 2006 in Oak Brook, Illinois.  The program featured a variety of speakers, vendor presentations, and break out sessions for discussion of local connectivity issues.   There was particular interest in preparation of Continuity of Care Record (CCR) by family practitioners and internists for generation of data in searchable fields for review in a web-based application. 

NOVEMBER 25, 2005.  Web-based Command!™ and updated versions of SpeechMax™ and SpeechServers™ are scheduled for release in January.  Command! for the Web will include web services with browser-based client for end user and administrator use.  As before, the software will permit multiple system, group, and user configurations.   The new program may be used with the old Command! to create a system of job transfer supporting email, FTP, and/or web services.  SpeechMax has been updated to support manual transcription and saving verbatim training files for creation or update of a speech user profile.  This permits pretraining of the speech recognition system prior to use.  The software supports multiwindow text comparison for speech recognition and/or manual transcription and Unicode for multilingual transcription.  The text editor is designed to support Dragon, support session files from a variety of speech engines.  Other features include built-in audio playback for transcriptionist foot control, text expander, customizable spell check, and audio and text annotations.  SpeechServers is being updated to run with the Microsoft speech engine.

JULY 16, 2005.  Market Partner Medical I.T. Pty. Ltd. in Queensland reports a novel approach to using Command!™ at a pilot project at a three-site clinic and outpatient center with Command!™ and other applications, principally Dox Document Management System, running centrally on a terminal server.  The three separate locations have a MPLS link between the sites replacing their former ISDN 128kbps links.  The low link speed was normally not an issue as all applications (Dox, appointment book, billing, MS Office, email, etc) all run on Windows 2003 server running terminal services.  However, microphone recording via terminal server was not an option without going to the expense of Citrix or other solution.  Command!™ again proved its versatility and that the major difference between the name-brand, $100,000 workflow terminal services solution and the inexpensive Command!™ product is usually just that--a $100,000.

JULY 8, 2005.  Custom Speech USA™ recently received certification as a Microsoft Certified Partner.  Certification was based upon independent software developer competency after VeriTest evaluation of Command!™.   Technical and marketing benefits are available to the Certified Partner.

MAY 14, 2005.  Command!™ web-based updates call for adding Command! web services (CWS) with centralized database with Windows web clients (WWC) and Command! web application (CWA).  The web application is planned as a web-based GUI operating on Internet Explorer or other browsers supporting file transfer from Windows, Apple, Linux, and other operating systems.  Functionality of previous solution would be maintained while adding new web-based options.  Comments or questions may be sent to info@customspeechusa.com

JANUARY 21, 2005.  Shopping Cart has been introduced for products and services.  Click here to view product selection page.

DECEMBER 31, 2004.  The company announced that it has over 500 licensees in the U.S., Canada, South America, Europe, Middle East, India, Australia, and the Far East, an increase of over 25% over last year. Resellers are located in the U.S. and several foreign countries.   

NOVEMBER 10, 2004.  Market Partner Julie Weight with AlphaBest has announced the creation of a users forum for Command!.  Based upon Julie's email, "The goal of the forum is to share information between users so that they have better experience with the products and possibly get some more ideas on how they can expand the application.  VARS are welcome to participate but sales pitches will not be allowed."  To access the forum, go here:  http://www.talkdocs.com/cgi-bin/blah/Blah.pl 

OCTOBER 31, 2004.  Palm Pilot technology is commonly used in health care, law, public safety, and other industries.  Custom Speech USA offers Palm OS software sound recorder from Audacity Audio, and other accessories for Palm dictation. This may be integrated with Command!™ workflow manager.  Audacity Audio recently added capability to permit transfer of dictation audio file from a Palm Treo 600 Smartphone.  This is described in Custom Speech USA eNews.

AUGUST 21, 2004.  acWAVE™ is a popular desktop tool for converting audio from Sony and Olympus handheld recorders. See October 2003 newsletter.  There is also SpeechPreferred™ text-to-speech desktop add-on.  The company provides US, UK, and Indian English, Spanish, Italian, Parisian French, and German voices from AT&T Natural Voices.  Excellent quality male and female voices are available from NeoSpeech for US English, Japanese, Korean, and Mandarin Chinese (female only).

JULY 28, 2004.  Market Partner Medical I.T. Pty. Ltd. was invited by its state government to exhibit at the forthcoming Health Informatics Society international conference being held in Brisbane (July 25-27). The state government is running under a theme of ‘Queensland e-Health Smart Solutions’.  A listing of exhibitors and sponsors includes Lanier Voice and Voice Perfect Systems, along with GE Healthcare, IBM, PeopleSoft, Hewlett Packard, Cerner Corporation, MEDITECH, and other health care companies. Medical I.T. reports that it will be showing a full implementation of Custom Speech USA products from auto submission of the dictation from its document management system to Command!, "dual-engine" editing with SpeechMax, approval, publishing the final document to the referring doctor, and archiving of the document files.

FEBRUARY 22, 2004. Custom Speech USA is completing a pilot project with independent pathologists near Spokane, WA.  The group needed a digital dictation system that would run "hands-free" in the laboratory.  They selected sound recorder CustomMike for use with Sennheiser microphones.  The sound recorder will be controlled with a standard, three-position foot control commonly used by transcriptionists.

DECEMBER 16, 2003.  A number of major changes to Custom Speech USA software have been made.  These include support of patch updates for all Custom Speech USA software, creation of searchable fields for subfields (UDAs) in Command!™ workflow manager, and integration of bar code scanning with CustomMike™ sound recorder for rapid, automatic dictation job creation and completion.  Scheduled release is mid-January 2004.  Software development kits (SDK) are now available for Command! and acWAVE.  Contact Custom Speech USA for details.

AUGUST 27, 2003.  A $100 manufacturer's rebate is available for purchases of the Command!™ Call Center (reg. $999) made by October 15, 2003.  Version 2 of the Call Center has recently been released and includes new user configurable modules for telephone dictation, voice mail, call forwarding, paging, informational announcements, and customer relations management.  The system can be used in medical, legal, insurance, public safety, and other businesses.  It may be fully integrated with the Command!! workflow system and Custom Speech USA for advanced speech processing applications, including speech recognition for Dragon and IBM speech engines, text-to-speech with AT&T Natural Voices, and more.  The system was recently reviewed in TechnoLawyer, an internet newsletter. TechnoLawyer--Command!! Call Center (08/03). 

AUGUST 8, 2003.  In some cases, Command!™ database constraints have been deleted during the creation of databases. These missing constraints will allow general use of Command! to possibly corrupt the database.  We are offering a free evaluation of your current database to all Command! users. Please contact via email at support@customspeechusa.com or by telephone at 219-662-3800 during regular business hours: M-F 9-5, CST excluding holidays. 

JUNE 15, 2003.  Customizable splash screen option.  Create splash screen for your company or customer that opens with Command!™.  One-time charge of $295.  Use same splash screen for multiple installations. Graphical editor for PathPerfect™ workflow designer.  Install as a standalone separate from Command!™.  End user can indicate suggested changes to workflows by modifying flow diagram.  Requires complete update, including database, from prior Command!™ versions.  FREE with purchase of Command!™ Professional or Enterprise.  Otherwise $49.

JUNE 1, 2003.  More flexible configuration options are available, including user specific prompts that can mimic prompts from Dictaphone, Lanier, or legacy systems for the Command!™ Call Center.  New "pause" function for telephone dictation that tells dictating user that system is "paused," not recording.  Voice prompts can also be created automatically using SpeechPreferred™ text-to-speech utility that comes FREE with acWAVE™.  Choose from multiple voice fonts in English and foreign languages from award-winning ATT Natural Voices.  Each voice font for ATT Natural Voices is $49.  Line tap (two-way recording) of phone conversations now included.  Record client or interviews and put on workflow.

MAY 20, 2003.  Stedman's Medical/Pharmaceutical Spellchecker 2003 is "high-speed medical spellchecking made easy!"  Produced by Lippincott Williams & Wilkins*, this best-selling spellchecker contains nearly half a million up-to-date medical, pharmaceutical, and bioscience terms.  It includes all new trade and generic drug names approved in 2002 including drugs from the Facts and Comparisons American Drug Index 2003.  The spellchecker includes current terms related to diseases, treatments, medical procedures, lab tests, medical and surgical equipment, eponyms, abbreviations, acronyms, and more from over 60 medical specialties.  The spellchecker add-on is available only for use with SpeechMax™ and will make it an even more valuable tool for transcriptionists.  Add-on price is $49.        

*About Lippincott Williams & Wilkins (www.LWW.com)

Lippincott Williams & Wilkins is a leading international publisher of professional health information for physicians, nurses, clinicians and students. LWW provides essential information for healthcare professionals in print and electronic formats. LWW is part of Wolters Kluwer Health, a leading provider of information for professionals and students in medicine, nursing, allied health, pharmacy, and the pharmaceutical industry.  Major brands include traditional publishers of medical and drug reference tools and textbooks, such as Lippincott Williams & Wilkins and Facts and Comparisons; electronic information providers, such as Ovid Technologies and Medi-Span; and pharmaceutical information provider Adis International. www.wolterskluwer.com

MAY 20, 2003.  With Command!™, use workflow for rapid entry of key word data into forms fillers using optical character recognition (OCR forms), such as OmniForms, or standard text processor, such as Word or WordPerfect.  Complete forms from voice data without the hassle or expense of traditional document assembly software.  Other product updates are also available: 

Command!™ Call Center:  new version with more flexible configuration options, including user specific prompts that can mimic prompts from Dictaphone, Lanier, or legacy systems.  New "pause" function for telephone dictation that tells dictating user that system is "paused," not recording.  Also available--voice prompts automatically created with text-to-speech.  

acWAVE™ and PlayBax™:  now convert VoiceIt .sri files, newer Sony .msv format, and Dictaphone Walkabout .dss to .wav. Batch processor (command line) version and acWAVE Software Development Kit now available. 

SpeechMax™:  new secondary session (.sx2) file for verbatim and transcribed text.  Use "session" file to select verbatim or final text and play back associated audio.  Medical spell checker and text expander available.    

MARCH 29, 2003.  SpeechMax™ can now compare speech recognition text with preexisting text form.  A "copy and paste" feature permits rapid copying of speech recognition text into the text form.  Post processing scripts can be created that add entries to a database, or format the final report or document within a transcription "shell."  The feature also permits comparison of speech recognition text with an old, transcribed dictation--a useful feature where dictation boilerplate is frequently repeated.  The implementation is available only for single engine processing. Other software updates include:  

Command!™ install kit includes generic templates for common workflows and activities.  These include workflows for dictation and transcription, speech recognition using Dragon NaturallySpeaking and/or IBM ViaVoice, and automated upload of audio from the Olympus DS 3000 handheld recorder.  A workflow specifically designed for radiology and pathology dictation and transcription workflows is available for customer demos. 

acWAVE™ and PlayBax™ now convert the newer Sony .msv format to .wav.  An acWAVE™ Software Development Kit is now available to Market Partners to develop similar functionality present in acWAVE, but with a customized front-end graphical user interface. 

MARCH 10, 2003.  IBM hosted its annual conference for IBM speech recognition dealers in Boca Raton, FL, on February 23-25.  Custom Speech USA gave a keynote presentation on its "dual-engine" technique for microphone, telephone, and handheld recorder dictation.  Automated transcription and training with SpeechServers™, and rapid editing using the SpeechMax™ text processor, were highlighted.  Coordinator for the conference is Jim Cox, president of Crown International, a Custom Speech USA Market Partner.

JANUARY 6, 2003.  With Custom Speech USA server software, front-end user interface to configure workflows and manage database not provided.  No separate purchase of Command!™ is required for configuration of server software if user already has license for Command!™ Call Center, acWAVE™--Server Edition, SpeechServers™.

JANUARY 6, 2003.  According to Keith Benman in "Patenting Northwest Indiana's Products," in 2002, the region's businesses were part of a patent rush that has seen applications double in the last decade, increasing from 106,698 in 1991 to 183,975 in 2001.  As business has become more global and competitive, protection of intellectual property rights has become more common.  Of 132 patents for inventions in which Northwest Indiana and Illinois border residents participated, only a few companies had multiple patents, including "Custom Speech USA, in Crown Point, a designer of speech recognition software . . . "  The article appeared on the first page of the Business Section of the Times, January 5, 2003.  

DECEMBER 7, 2002.  Law Office Computing provides independent software reviews and technology review for the legal industry.  SpeechProfessional™ is reviewed in the December/January 2003 issue.  Grace Suarez, a sole practitioner and management consultant in San Francisco, found "many practical uses for SpeechProfessional.  Aside from its use in the office, an attorney who is on the road a lot can send his or her recorded dictations to an assistant.  The assistant can then use SpeechProfessional, and send the transcribed product back to the attorney, or mail it directly to the client."  She concluded, "Until voice-to-text programs become 100 percent accurate, a solution such as this, which puts a human transcriber between the audio file and the final product, is a smart idea.  SpeechProfessional does 95 percent of the transcriber's work.  The only thing left for the transcriber to do is make sure the end product is accurate."  She recommended it to attorneys who wanted to increase their productivity and have support staff available to transcribe.  Click here to see review.

NOVEMBER 16, 2002.  LegalTech Chicago 2002 was held at Sheraton Chicago Hotel and Towers Cityfront on November 4-5.  At the booth, visitors could see a demo of SpeechProfessional™, the revolutionary software from Custom Speech USA™ for use with Dragon NaturallySpeaking and IBM ViaVoice to reduce correction time and automate speech recognition training.  LegalTech shows are designed to help visitors identify new, cost-saving technology for the legal industry.  Based on number
of booth attendees for demos, Custom Speech USA™ was the most popular booth or show visitors.  A downloadable, interactive demo showing the "dual-engine" technique with legal dictation is available on the company website.

OCTOBER 17, 2002.  Law Technology News is one of the leading technology journals for the legal industry.  The September issue contained a review of the dual-engine approach and the SpeechProfessional™ combo package. 

SEPTEMBER, 14, 2002.  Custom Speech USA was nominated for an award in the Automatic Speech Recognition (ASR) competition sponsored by the prestigious Speech Technology Magazine.  Nominees included Microsoft, IBM, ScanSoft, and other companies.  To cast your vote for Custom Speech USA, go to www.customspeechusa.com and click on the blinking button SPEECH SOLUTIONS AWARDS.  On the Speech Technology site, go to the second drop down Best ASR Solution/Desktop and select Custom Speech USA.  
PRESS RELEASE

SEPTEMBER 1, 2002.  Beginning September 1, new software products and services are available from Custom Speech USA™ or Authorized Resellers:    

New Products:

• Forms Transfer for PDA $199
• Web Reports $199
• Document Processor Server $199
• File Monitor Server $149
• Telephone Dictation Server Combo $1,199 (SAVE $148)
• SpeechProfessional™ Introductory Package $549 (now includes a FREE copy of  File Monitor Server--a $149 value)
• SpeechPreferred™--text-to-speech functionality for acWAVE™
• Drivers for Philips SpeechMike series--now included in CustomMike™

Forms transfer for PDA is a Command!™ add-in that permits custom creation of forms for upload and download of data to and from a Palm Pilot into an the Command!™ system.  The software is available by special order and requires custom creation of interface between the Command!™ database and an Access database for each PDA form.  Point of service data entry with a Palm Pilot may be of assistance in health care, law enforcement, retail, and many other areas.  Form data may be easily uploaded and downloaded into the Command!™ file management, document retrieval, and time tracking system with customized forms created by Custom Speech USA™.   One potential application is point of care interventions by pharmacists. 

Web Reports is a Command!™ add-in that permits access to Command!™ Reports add-in via intranet or internet.  Logon requires Command!™ User ID and Password.  Now your customers, suppliers, or vendors can have web access to selected job information.  Decrease unnecessary phone calls, faxes, or emails by making selected, password protected workflow information available to your customers or business partners on their web browsers. For example, a recent accounting installation now has the ability to provide user ID and password-protected web information to its clients about additional financial information required to complete tax forms.

Document Processor Server permits post-processing of text files for insertion or deletion of text using search and replace functionality.  Use to replace specified trigger word or phrase with boilerplate text.  For example, the term "normal chest" might be replaced with the text for a normal chest x-ray report ("Heart size is normal.  Lungs appear clear of active infiltrates.  Impression:  No evidence of active lung disease.")  Data extraction is also available to determine incidence of selected terms and expressions for data mining.

File Monitor Server periodically queries a selected folder for specified files.  These are transferred to a Command!™ workflow for processing.  For example, files downloaded from an Olympus DS 3000 into the Olympus software folder may be automatically transferred by File Monitor to SpeechServers™ for automated transcription and then to SpeechMax™ for review and edit.  File Monitor may be used with audio and other file types. 

Telephone Dictation Server Combo includes Telephone Dictation Server (reg. $999), Command!™ Standard ($249), and acWAVE™ Server Edition (reg. $99).  This is a complete package for the customer that does not already have Command!™ file management or acWAVE™ audio file conversion.  Great starter kit for a small office or small, independent transcriptionist.

SpeechProfessional™ combo includes SpeechMax™, for rapid edit of text transcribed using Dragon and IBM speech engines.  More than a text processor, SpeechMax™ highlights likely errors and plays back selected audio.  SpeechServers™ transcribes dictation audio using Dragon and IBM engines, and automatically helps train your speech recognition system.  SpeechServers™ Introductory Package has a 3 million word limit.  Refill license packs are available.  The SpeechProfessional™ suite also includes Command!™ workflow manager, acWAVE™ audio file conversion, and File Monitor Server.  For example, with virtually a single mouse click, you can extract .dss files from an Olympus folder, convert the audio to proper .wav format, send audio to SpeechServers™ for batch, automated transcription by Dragon and IBM, and return the text to SpeechMax™ for edit and review.  And this can all be done on a single computer!  A stunning, revolutionary package for home office or large enterprise that makes speech recognition work the way it should.  Speech recognition engines not included.   

SpeechPreferred™ may be used with SAPI 5 compliant text-to-speech engines.  Distributed with Microsoft engine.  May use with others, such as ATT Natural Voices. 

New Drivers for CustomMike™ have been created for the Philips SpeechMike series.  These microphones uses serial or USB ports and may be used with the CustomMike™ software sound recorder.

New Maintenance Services:

• Bronze Maintenance $147/yr (3 hrs)
• Silver Maintenance $269/yr (6 hrs)
• Gold Maintenance $359/yr (9 hrs)
• Platinum Maintenance $699/yr (20 hrs)

Maintenance Plans include support for basic configuration, and not advanced scripting or programming.  It is available Monday through Friday, 9 AM - 5 PM, Central Standard Time, holidays excluded (New Year's, Memorial Day, July 4th, Labor Day, Thanksgiving, and Christmas).  Time not used on a support plan during the one year subscription cannot be carried forward.  Use these maintenance plans to let Custom Speech USA™ create a customized, factory-configured workflow solution that you can download from the internet.  Save time and money by providing user names, activity, and workflow information to Custom Speech USA™, and letting its highly qualified personnel configure your system for you.  Excellent examples of this "factory-configured" approach include recent health care and law customers.

AUGUST 24, 2002.   Authorized resellers from New Jersey, Connecticut, Mississippi,
Michigan, Minnesota, Indiana, Illinois, Oklahoma, and California attended a two-day
training course in Crown Point about the company's software applications for dictation,
transcription, speech recognition, and office workflow automation. The company has
58 authorized resellers in the United States, Canada, Australia, New Zealand, and Germany.

JULY 9, 2002.  "The off-the-shelf products require doctors to act as their own typists and they don't have time to do that.  This makes IBM and Dragon products better."  This quote from Ted Rosdil, business manager, was the lead-in to today's article "Seeing What You Say," by Brian McCollum, business staff writer, on the front page of the Business Section of The Times, published in Northwest Indiana.  The article explained that the company's software is designed to improve currently available off-the-shelf speech recognition software.  "The software has traditionally been used by the medical community, enabling users to talk to a computer so notes and other information can be transcribed quickly.  However, some area medical professionals said the software causes more problems than it solves because there were too many mistakes which often had to be corrected manually, taking up valuable time.  That's where Custom Speech USA plans to improve the market."  Developed over a 5 year period for use with Dragon NaturallySpeaking and IBM ViaVoice, the software is targeted at a wide variety of businesses and professions. Chuck Runquist, a New Jersey-based independent technical consultant who formerly worked as a product manager with Dragon NaturallySpeaking, noted, "This product will fill a much-needed niche.  Many doctors and lawyers don't want to go through the complex process of registering their voice which is what the other products require.  Custom Speech USA has removed the tediousness of it.  It allows anyone to edit, modify, and produce accurate text.  Rosdil said the company was just awarded a contract with the Los Angeles Coroner's office, and one of their software designers will be going there to install it." Click here to see the entire article.    

JULY 4, 2002.  The Post-Tribune featured Custom Speech USA™ on the front page
of its Business section in "Talking to Software: Crown Point Firm's Products are
Closing this Communication Gap." The article was based on an interview with Ted Rosdil,
the company's business manager. Noting that the software had taken five years to develop
and was introduced two months ago, Bob Craig, Post-Tribune correspondent, wrote: 
"In the modern world of computers and technology, speech recognition software offers
the promise of hands-free computing use and a glimpse into the future.  That glimpse
has been blurry, though, because of the difficulty in training software to accurately recognize
user voice patterns and create error-free text from the spoken word.  Custom Speech
USA, a Crown Point software firm, is creating products that bridge the gaps in off-the-shelf
speech recognition software and, the process, revolutionizing application of the technology. 
The first's software products are designed as add-ons to popular speech-recognition software
such as Dragon NaturallySpeaking and IBM ViaVoice.  Marketing the products to
medical and legal professionals who dictate recorded notes on a regular basis, the firm
has already landed the Los Angeles coroner's office as a client."  Rosdil, the company's
business manager explained, "We've finally come up with a product that allows a busy
professional to use speech recognition software the same way they use transcription." 
The products are the result of the efforts to make speech recognition more user friendly
and ultimately more widely used.  "The off-the-shelf products are designed for real-time
dictation into a personal computer.  As the user speaks, the software transcribes--with
the results appearing on the monitor. In the medical and legal fields, the dictation is
often done with a hand-held recorder, then transcribed by a transcription professional later. 
On Custom Speech USA's server-based products, the computer transcribes the information,
whether it has heard your voice before or not."  SpeechMax, one of Custom Speech
USA's products, works like a word processor, and uses the text from both the Dragon
and IBM programs to highlight probable errors, thereby enabling editing to be accomplished
more quickly.  The corrected text is resubmitted to the Custom Speech USA servers for
"speech training," that permits the  programs to 'learn' from the corrections.  As
the software becomes more accurate, less time can be spent correcting text using
SpeechMax.  Quoting Rosdil, "'It's a revolutionary kind of software. It's inexpensive,
simple to implement, and it uses off-the-shelf software. We're not reinventing the
wheel here.'" Located in downtown Crown Point, the products are being marketed
through a network of licensed distributors. "'The software is quite powerful and requires
a certain amount of training, and the person who installs it really needs to understand it.
We have 40 resellers and inquiries daily from new people. The growth has been phenomenal.'"

JUNE 22, 2002.  Custom Speech USA™ software was demonstrated at the 11th
Annual Conference of the New Hampshire Medical Group Management Association
at the Inns at Mills Falls in Meredith, New Hampshire on June 19-21.  NHMGMA is
the Granite State’s nonprofit association of more than 300 managers of health care
groups.  This year's conference theme was "Survival Skills for Health Care Managers."
The Custom Speech USA™ speech processing applications were showcased by
Thomas Westheimer, Compucare, Inc. (Petersboro, NH).  According to reports,
the SpeechMax™ demonstration was one of the more popular demonstrations.
Compucare is one of the more than 40 authorized resellers of the company's
products in the United States, Canada, and Australia.

JUNE 20, 2002. "Crown Point Firm Takes Lead in Voice Recognition," was the
headline of a recent article in the Crown Point Star, published locally since 1857. 
The article began:  "Imagine you're driving home and suddenly you need to send a
message to a family member, or to an employee or your doctor.  If you have a cell
phone and the proper connection, you can simply dictate while you drive and message
will appear where you want it to appear, be it on a computer screen, in your own
computer or on someone else's message screen.  The imagining is complete. 
The reality is here.  People all over the world can now dictate to their computers,
through spoken words, and have the dictation immediately transcribed and saved. 
The genesis of this next off-ramp on the computer superhighway started right here,
at 134 S. Main St., in the shadow of the of the magnificent old Lake County Court
House."  The article further described the company's products and potential uses
in the health care, public safety, and other areas. Ted Rosdil, the company's manager,
noted that the company's computers "can handle languages spanning English, Japanese,
Portuguese, German, French, and Spanish.  So there is no international barrier to
the system."

JUNE 7, 2002.  Chuck Runquist, former technical project manager at Lernout & Hauspie,
is an Authorized Reseller and consultant for Custom Speech USA™.  In response to a
question on an internet voice users group whether the "dual-engine" technique using
the Dragon NaturallySpeaking and IBM ViaVoice engines would result in less accuracy,
Chuck wrote:  

Since I left L&H/ScanSoft in December, I have been working with Custom Speech USA.  

First, all the applications developed by Custom Speech USA are unique in the industry and provide solutions which are available now that I had intended to include in the Dragon NaturallySpeaking® version 6 SDK/Developer Suite product.  This includes on-the-fly transcription without the necessity to pre-enroll a user, taking dictation over the phone and converting it to the proper Dragon format and automatically transcribing it, and many other significant applications that make effective use of the speech engines and automated workflow.

Contrary to your suspicions, using the dual engine approach for transcription not only improves accuracy but can be coupled with a very sophisticated editing and training process which updates the user for both engines transparently and on-the-fly each time it is run.  I could go on for another page and a half explaining how it works but I think that that would generate more questions than answers.  Suffice it to say that this process has been tested thoroughly by Custom Speech USA with the result being that the combination of the two engines combined (and they are the top two engines which is why it works so well) has been shown to statistically demonstrate that where the two engines agree, the transcribed text is virtually 100% accurate.  Where the transcription is different between the two engines, their studies have shown, and I tend to agree having used it and understand how it works, almost 99% of the time one or the other transcribe the dictation correctly.  By using a patent pending compression algorithm, Custom Speech USA is able to map the differences, display them color-coded, and provide a very fast and efficient way of making corrections by concentrating primarily on the differences between the transcription provided by the two engines.  The process can then be run through their SpeechMax utility, which allows you to very quickly highlight the differences and playback the audio behind each.  In addition, the amount of text (by setting the timing) read back can be can be set by the user such that the playback function can begin slightly before and slightly after the selecting text, thus giving the user the ability to adjust the playback so as to be able to provide more audio context when and if necessary. Further, the process of editing, using the tested assumption that if both engines transcribe a given word, phrase, sentence, or selection identically, then there is a better than 99% probability that it was transcribed correctly by both engines, the process of editing the transcription is faster, easier, and 10 times more efficient than normal transcription.

In addition, there are some assumptions made when looking at the normal transcription processing Dragon NaturallySpeaking® that are based on the fact that the developers wanted to ensure that the average user does not go off on their own in testing transcription processes.  Over the course of the last six months I have been testing various methodologies for doing transcription based on the underlying assumption behind the Custom Speech USA approach.  I would be glad to discuss my findings with anyone on a one-to-one basis by phone.  It is too involved in to engage in a discussion via e-mail.  Regardless, what Custom Speech USA does using the dual engine TurboTranscribe method results in extremely high accuracy combined with ease-of-use.

Lastly, the approach the Custom Speech USA takes does not require a user to pre-enroll.  A new user, even without a profile created and enrolled, can use the Command! Call Center, call in and dictate over the phone, and the Custom Speech USA product will transcribe the dictation, the allow the editor or transcription is too quickly make corrections, and then automatically used the corrected text and audio to create and train a user on-the-fly.

I can tell you this, properly set up and utilized the accuracy and the ease-of-use of this product will knock your socks off.  One demonstration that we use is to allow potential VARs and customers call into the call center, dictate a couple paragraph, and then we send the transcribed text back to them via e-mail.  In this process there is no currently enrolled user, and the results are generally very impressive.  This entire process takes about two minutes to complete once the tester has hung up the phone. Custom Speech USA also has done an impressive job in maintaining the quality of dictation received over the telephone (8 MHz by 8-bit sampling rate --11.05 MHz by 16-bit Dragon sampling rate conversion process).

The Custom Speech USA programmers are thoroughly knowledgeable and technically skilled using the Dragon NaturallySpeaking® SDK.  In fact, they are more knowledgeable than 99% of the developers with whom I worked while I was with L & H regarding the use of the SDK, and have fully speech enabled all of the Custom Speech USA products.  I can only say that I have worked extensively with them on learning the underlying functionality of these products and I am thoroughly impressed with their knowledge and their skill, as well as what they have created as far as these products are concerned.

I would be more than happy to discuss any of this with anyone, as well as provide greater detail and explanation.  Since there is a lot of information to cover, I would suggest that interested parties contact me off line and I will provide you with my phone number.

With my reputation on the line, as well as my knowledge and understanding of programming, I would not be sharing this information with few if I had not seen it, used it, and if I were not thoroughly impressed with Custom Speech USA's applications.

Chuck Runquist

This document was dictated using Dragon NaturallySpeaking® version 6 

JUNE 5, 2002.  TechnoLawyer Community (www.technolawyer.com) provides an internet
newsletter for over 6,500 attorneys. SpeechProfessional™ and other Custom Speech
USA™ products were reviewed in today's TechnoLawyer Community NewsWire.  The
article was entitled "Two Speech Recognition Programs are Better Than One," and
was written by Jennifer Pascoe (jenniferpascoe@earthlink.net):  "Thanks to Custom
Speech USA's SpeechMax, the days of having to choose between Dragon NaturallySpeaking
and IBM ViaVoice have come to a close. Operating on the two-heads-are-better-than-one
principle, a companion application called SpeechServers runs your dictation through
both NaturallySpeaking and ViaVoice. SpeechMax then compares the two results,
and enables you to correct the text far more rapidly than you could when using
NaturallySpeaking or ViaVoice alone. SpeechMax can display a split screen containing
the transcription from each program, and you can quickly select the text from Dragon
or IBM that is correct for the final version. To speed up the correction process,
SpeechMax's 'TurboTranscribe' technology highlights the likely errors, and enables
you to playback specific portions of the original audio by simply selecting text. (Using
the optional VerbatiMAX technology, you can also compare manually transcribed text
to speech recognition text to generate verbatim text for automated speech training.)
After correcting the text, you can send it and the accompanying audio back to SpeechServers
for automated, repetitive training to further improve recognition accuracy. SpeechMax
comes in four different versions (Basic, Advanced, Preferred, and Trainer). Prices start
at $99 for the Basic version. Also available is the 'SpeechProfessional Suite,' a bundle
that includes both SpeechMax and SpeechServers as well as Command!, a workflow
management tool, and acWAVE, an audio file conversion utility." TechnoLawyer is
a popular legal technology and practice management resource with over 15,000
subscribers.  It consists of a network of free, critically-acclaimed
e-mail newsletters, and a searchable Web-based repository of all TechnoLawyer
content since January 1997.    

JUNE 1, 2002.   Authorized resellers from Florida, Virginia, Maryland, New Jersey,
Michigan, Indiana, Illinois, Wisconsin, Missouri, Montana, and California attended a
two-day training course in Crown Point about the company's software applications for
dictation, transcription, speech recognition, and office workflow automation. The
company has 37 authorized resellers in 19 states and Canada. 
PRESS RELEASE

MAY 1, 2002.  Custom Speech USA™ participated in a panel discussion "Where Are
We Today with Speech Recognition?" at the 13th Annual Medical Transcription
Industry Alliance meeting at the Hyatt Regency Tampa Hotel, Tampa, FL, April 25-26,
2002.  Other panel participants included representatives from DVI, Speech Machines,
and MD Productivity.  Discussion focused on audio processing and training of speech
files.  The Custom Speech USA™ presentation focused on SpeechMax™--more than
a word processor, it helps you rapidly edit text using speech recognition.  Use of
SpeechMax™ to produce final text and create verbatim text files for automated speech
training was demonstrated.  The company distributed flyers "Revolutionary New Software
for Practical Solutions".  Ted Rosdil, Business Manager, and Karen Claus, Office
Manager, explained in some detail to the many booth visitors the significant advances
that the company had achieved in workflow management, computer telephony, and
dictation, transcription, and speech recognition.

MARCH 28, 2002.  Custom Speech USA™ announced the release of Custom Speech
USA Complete, a single CD with software for workflow automation, computer telephony,
and speech processing applications. A 30-day trial version is available for purchase. 
PRESS RELEASE

FEBRUARY 2, 2002. Custom Speech USA™ installed Command! for an accounting
firm in Michigan City, IN. The accounting group uses Command! as a file manager to
organize workflow for the firm's record sheets for individual income tax forms.  

DECEMBER 19, 2001.  Authorized Reseller A.E.C. Solutions, Inc. installed Command!™
and Command!™ Call Center for a medical transcription service in Oklahoma City,
Oklahoma.  The service has 15 at-home transcriptionists and provides medical
transcription for a 250-bed local community hospital.

JUNE 4, 2000.   The OPEN MRI 2000 Conference was held May 31 - June 4, 2000
in Las Vegas, Nevada. The conference was attended by numerous radiologists,
administrators, and technologists associated primarily with low field MRI systems.  
Conference chairmen were Peter A. Rothschild, M.D., Medical Director, Open MRI
and President, Image Enhancement System, Roger D. Freimarck, R.T., Northwest
Imaging Forums, and Nelson Acosta, CEO, The OMI Group.  Participating low field
MRI manufacturers included GE, Hitachi, Marconi, Siemens, and Toshiba.  Custom
Speech USA, Inc. products were displayed at a booth and described in a conference
paper entitled "Cost Effective Voice Recognition Dictation Systems."
PRESS RELEASE

DECEMBER 3, 1999.  Radiology Society of North America 85th Scientific Assembly
and Annual Meeting was held at McCormick Place in Chicago on November 28 -
December 3, 1999.  Attendance was again estimated at over 60,000.   Visitors to
the CSUSA booth were shown the new product TurboTranscribe™, a method for
combining speech recognition engines for more rapid generation and editing of transcribed documents.  Participating vendors included Thomas Westheimer, Compucare, Inc.
(Petersboro, NH) and Ms. Barbara Grow, host of MT Meeting Place™.

DECEMBER 1, 1999. In his email newsletter Computers for Lawyers, Paul Bernstein
noted that every lawyer wanted to use Dragon NaturallySpeaking or similar program
to dictate or control his or her computer with voice commands. Many lawyers had
tried the technology, but advised that "it does take time and tenacity to get there.
I too have tried Dragon and loved it and I do know that if I gave it enough time and my
 effort, I too would be dictating all my newsletter, pleadings, and correspondence with
it . . . but, to date, I have not had that kind of patience." Bernstein explained that about
a year previously he had seen Custom Speech USA™ products at the ABA Tech Show.
This software would allow lawyers to "train" Dragon NaturallySpeaking and other speech
recognition products "in a fraction of the time, and it sure looks like they have
something now. I talked with the owners and hope to have the opportunity to personally
with the product on my own computer."

NOVEMBER 15, 1999.  Custom Speech USA™ attended the LegalTech show that
was held at the Sheraton Hotel in Chicago.  According to show statistics, more
attendees visited the Custom Speech USA™ booth than any other.  We demonstrated
a new program called SpeechTrainer™.  The program provides a user friendly interface
that permits your secretary to take your daily dictation and help train your speech
recognition system.  We also introduced another program called Command!™ that
integrates speech recognition with desktop business applications such as sound
recorder, word processing, speech recognition, automated translation, handwriting
recognition, document and file storage, fax, and email.  Command!™ provides easy
access to audio and text files from day-to-day dictation for automated speech training.  
Custom Speech USA™ will be attending the RSNA show at McCormick Place in
Chicago November 28-December 2.
PRESS RELEASE

MARCH 20, 1999.  ABA TechShow™ 99, Sheraton Chicago Hotel & Towers, Chicago.  
This 3-day symposium was presented by the ABA Law Practice Management Section.   
Over 200 lawyers, legal secretaries, and administrators visited the Custom Speech USA, Inc.
booth.  The prototype for the standalone version of the Custom Speech USA™ SpeechTrainer™
was demonstrated.  With this software, a secretary can help train the system for a
dictating lawyer or doctor.  The prototype version uses Dragon NaturallySpeaking
Professional Edition.  Updates are planned for use with IBM, L&H, and Philips speech
recognition systems.

Price, terms, specifications, and availability are subject to change without notice. Custom Speech USA, Inc.
trademarks are indicated. Other marks are the property of their respective owners.