Masthead.jpeg (68003 bytes)

Number 53, August 2006

VERSION 9 DRAGON SPEECH RECOGNITION--SPECIAL OFFER

Product prices for version 9 are Preferred $199, Professional $899.99, Medical $1,199.99, and Legal $1,199.99. Offer expires 9/15/06, upgrade to version 9 for only $99.99 for $99.99 for Professional, Medical, or Legal.  The special upgrade is only available for box customers when upgrading from prior edition (for example: Professional 8 to Professional 9).  A flat $9.95 shipping and handling fee will be added to all orders for shipments within the continental U.S.

WEB SERVICES FOR DIGITAL DICTATION AND TRANSCRIPTION

 Web services are now included with Command!™, and are available as a FREE upgrade to preexisting customers.  The new feature was recently implemented in an Australian multisite medical practice by Custom Speech USA reseller Medical I.T. Pty. Ltd. There was no significant workflow change of service interruption for the dictating physicians.  According to one of the reseller's staff, "The typing ladies think the new server is wonderful - they especially like being able to see the logs in a wide format and scroll through the log whilst messages are being downloaded."

INFORMATIONAL SERVICES WITH COMMAND!™ CALL CENTER FOR "311" CALLS

"Move over 911; the new hot government dial-code is 311. Municipalities across the country are rolling out the quick-dial government information line to ease the number of non-emergency calls to 911. The Federal Communications Commission recently mandated that all Internet-telephony services support the new standard, which will require extensive modifications to existing networks. There are also calls for creating other X-11 numbers for quick informational services . . . . " Government VAR, "Dialing for Services" (August 21, 2006) Here is an opportunity to use voice mail (message center), reporting and dictation, informational announcements, call transfer, and call management available with Command!™ Call Center.  Flexible and scalable, the system can handle hundreds of simultaneous calls or more, depending upon the one or more telephony cards selected.

SPEECHMAX™ NEW FEATURE--SPEECHCENSOR™

HIPAA and other confidentiality legislation limit disclosure of patient name and other identifying information. The transcription industry is seeking to upgrade  quality by instituting MT apprenticeship programs (see below) and improving on-the-job training.  But how can a transcription company supply material for novice transcriptionists as part of an MT apprenticeship training program, but delete sensitive patient information from training audio or text files?  With the new SpeechCensor™ feature in SpeechMax™ (U.S. and foreign pats., pats. pending), a user can simply select the word or phrase, select speech delete, and create a new audio file with a beep inserted in place of the deleted word or phrase.  In cases, where the audio tag is a longer segment, the integrated SpeechAnalysis™ tool can be used to resize the length of the audio tag, so that only limited audio, say the patient name, is censored, not the surrounding audio in the utterance.  This feature can be used with session files created from speech recognition, as well as segmented dictation audio that has been previously manually transcribed in SpeechMax™.  There are other potential uses.  For instance, SpeechCensor™ may also be used to help balance the needs of anti-terrorism vs. privacy rights by providing a tool for rapid deletion of confidential audio and text information before distribution to local law enforcement.  As indicated in a recent report,  "Nearly five years after 9/11, reports show the federal government has made limited progress on sharing terrorism information because of uncertainty about what to share, and how to do so without infringing on civil liberties."  A. Lipowicz, "To Be or Not to Tell:  Privacy Rights Vie with Anti-Terrorism Goals at Intel Centers," Washington Technology (July 24, 2006).

SPEECHMAX™ FOR CREATION OF VOICE NOTES

In an editorial dealing with speech recognition for dictation as a productivity tool, summarized below, publisher Bill Meisel has indicated that the greatest contribution of speech recognition may be for "idea capture" or "voice notes." In recently released, updated SpeechMax™ (U.S. and foreign pats., pats. pending), techniques for idea capture include (1) real-time, interactive speech recognition with the "plug-in" Dragon, IBM, and SAPI 5.x speech engines, or (2) integrated sound recorder and local or remote server-based speech recognition of the audio file by SpeechServers™ (U.S. and foreign pats., pats pending). The transcribed voice notes may be reviewed and corrected locally in SpeechMax™ by the dictating speaker, or  "delegated" for correction by a remote editor using SpeechMax™.  The author can also save the uncorrected audio-aligned, speech recognition text without review as a jog to memory.  If necessary, the speaker can select text in this "rough draft," and play back the audio for the exact word or phrase.   With SpeechMax™, a user may also  create voice or text comments in an integrated annotation window for text in the main read/write window.  This annotation may be associated to an adjacent digital photo or graphic that is displayed in the HTML editor, for instance, as part of an electronic scrapbook.  The annotation may be recorded with integrated sound recorder, text from keyboard entry or speech recognition, or combined voice and text.  The annotation may also be a path string to a website or a program executable.  SpeechMax™ also supports "multilevel" annotations, e.g., one or more annotations, by the same or different user, or multiple hyperlinks or program executables associated to the same text in the main read/write window.  Using SpeechMax™ text to speech with ATT Natural Voices, with voice fonts for U.S. English and many European languages, or NeoSpeech VoiceText for English, Mandarin Chinese, Japanese, and Korean, the user can convert text notes into speech. 

SPEECHMAX AND TRANSCRIPTIONIST QUALITY ASSURANCE AND TRAINING

SpeechMax™ provides many features that can help train new or apprentice transcriptionists, or provide continuing education for experienced personnel. Traditional text editors for manual transcription require the novice to use a footpedal to locate possibly mistranscribed dictation within the audio file. In SpeechMax™, the operator may select audio-aligned text that has been incorrectly transcribed and playback the audio immediately. Further, using text compare the apprentice can compare the original transcription and the corrected text prepared by QA personnel or instructor to quickly locate mistranscribed text. Because the application supports editing session files from several server-based speech recognition engines, it can also be used for speech recognition QA, as well as training transcriptionists to edit server-based speech recognition applications.  Unlike traditional manual transcription, with speech recognition a major issue is not incorrect spelling, but misrecognized words. And, with SpeechCensor™, as described above, a medical transcription service can create "censored" training material (without confidential information) for both manual transcription and speech recognition.    

SPEECHMAX™ AND THE EMERGING ROLE OF MTs AS MEDICAL EDITORS

It is generally believed that greater adoption of speech recognition, as described below, will result in transcriptionists correcting session files from server-based speech recognition.  SpeechMax™ supports playback of the audio-aligned text of the session file with a standard, three-position footpedal that is standard in the industry, e.g., Infinity transcriptionist footpedal, or other USB HID compliant device.  Plus, the application includes a built-in text expander, spell checker, basic editing and formatting, advanced text comparison to check output between two or more speech engine texts (as well as previous manually-transcribed reports), and training features for an MT apprenticeship program, as described above.  The session files from SpeechServers™ transcription with Dragon, IBM, Microsoft, and SAPI 5.x speech engines can also be edited in SpeechMax™. Further, the system supports editing Dragon (DRA) session files from off-the-shelf, runtime, and Dragon Server versions using a footpedal to playback Dragon (DRA) session file.  You can also export the session file audio as a WAV file for playback in PlayBax™ (or similar playback application), and transcribe into a standard word processor, such as Microsoft Word or Corel WordPerfect. The MT can also create text and audio annotations with an integrated annotation sound recorder and auxiliary text window, thereby providing messages for the reviewing physician. 

SPEECHMAX™ FOR FUN

As an offshoot of development for transcription and speech recognition, several features in SpeechMax™ support display of embedded multimedia.  As an HTML presentation editor,  SpeechMax™ graphics and images may be displayed against customizable backgrounds ("style sheets"). Text and audio annotations also may be recorded for digital photos, creating novel opportunities for electronic scrapbooking with personalized voice comments or music, synchronized playback of music and lyrics for singalongs, or playback of music alone with customized graphics and images for personalized karaoke.  The application also supports creation of AV text for audio books, lectures, or sales presentations. 

SPEECH RECOGNITION FOR DICTATION AND PRODUCTIVITY

Bill Meisel, publisher of Speech Strategy News, wrote recently that the major hurdle to broad adoption of speech recognition for general dictation is that "composing a finished document by voice is not a natural skill--it is acquired.  We typically composed documents by keyboard, and we converse with speech." He further argues that speech recognition "is not a good keyboard replacement unless one simply cannot use a keyboard. For example, it is much easier to edit with a keyboard and mouse than by voice." Further, speech recognition is excellent for "getting one's ideas into rough text without editing . . . . Dictation software for the business or consumer should be sold as 'idea capture' or 'voice notes' software to get one's ideas down efficiently--in essence, close to a 'though-to-text' tool and a new productivity category. Speech recognition excels at getting down a rough draft of ideas without temptation to constantly edit." B. Meisel, "Dictation as Productivity Software," Speech Strategy News (August 2006).

MT APPRENTICESHIP PROGRAM

The American Association for Medical Transcription (www.aamt.org), the Medical Transcription Industry Association (www.mtia.com), and Department of Labor -- Employment and Labor Administration (www.doleta.gov) are backing an apprentice program for medical transcriptionists. As explained by MTIA, "The apprenticeship model can be an effective system for addressing human resource issues and skill shortages by providing the expertise and knowledge individuals need to do their jobs effectively and advance in their careers. This model allows for flexible competency-based training that enables apprentices to move at their own pace, benchmark achievements, and build a portfolio of skills. The strategy is to create a continuum of career advancement opportunities including career credentialing." In one approach, an employer would hire an apprentice for 2 years who would receive training and instruction while earning increased wages. See also T. Cassidy, "The Apprentice--Medical Transcription's Big Summer Hit," (July 3, 2006)

MTs AS MEDICAL EDITORS WITH SERVER-BASED SPEECH RECOGNITION

In a two-part series on speech recognition technology (SRT) in Advance for Health Information Professionals, front-end, real-time, interactive speech recognition is described as requiring more physician interaction and training from the physician "which we all know docs could do without." Front-end dictation also requires the physician to dictate directly into a PC, thereby raising portability issues as well. Quoting one vendor, "'Back-end speech recognition lets physicians dictate as they always have: into a telephone, a portable digital voice recorder, a PDA, or even directly into a PC-based EHR [electronic health record] . . . . The voice file is run through the recognition engine in the background, without the physician even needing to know this is going on. The payoff is MTs can substantially improve their productivity by editing a draft report rather than having to transcribe it from scratch.'"  Are MTs accepting their new roles as speech recognition editors?  Some MTs are "incensed" and "against their associations and profession as a whole for lack of appropriate representation, standards, and certification--not to mention their fear of losing their jobs." Not only are job requirements different, many MTs finding themselves many are adjusting to the new title of medical editor (ME). "But, positive feedback has been given from current MEs who are utilizing back-end SRT for improved work conditions and productivity."  As one director of a large, nationally-based transcription service put it, "'Back-end SRT allows an ME to be more productive, less prone to injury and experience a more diverse and rewarding work environment.'" T. Cassidy, "Back-End Speech Recognition: Piece of Cake for Docs to Dictate," Advance for Health Information Professionals (July 17, 2006), T. Cassidy, "Medical Editors":  Transcriptionists Newest Role," Advance for Health Information Professionals (July 31, 2006)

COSTS FOR CREATION OF E-HEALTH RECORDS

Estimated cost to digitize public and private U.S. health records is $100 billion.  By comparison, total state and local government IT spending on health and welfare in 2006 is estimated at $7.6 billion. Currently, there are 28 states planning to implement e-health records systems, with 9 governors who, in their State of the State speeches, acknowledged the need for health IT improvements.  E. Butterfield, "Show Time! for e-Health Records" (Washington Technology, August 7, 2006). Perhaps due to high costs of implementation, a report funded by the federal Agency for Healthcare Research and Quality found that many "characterizations of progress [in the development of regional health information organizations or "RHIOs"] have been overstated and few projects are yet exchanging clinical data." D. Raths, "Real Solutions for RHIO Problems," Healthcare Informatics (August 2006). For these reasons, Custom Speech USA is continuing its ongoing research and development to develop cost-effective, affordable solutions for electronic health records and other e-health record systems that integrate preexisting software and web-based interfaces, not generation of custom interfaces that increase the costs for patients, doctors, and hospitals.  A particular area of interest is entry of dictation and transcription and speech recognition data into defined, searchable fields using the Command!™ web services workflow manager.  Once entered into a database, this information can be displayed on a secure web reporting tool.

Custom Speech USA, Inc.
3 North Court Street, Suite B365
Crown Point, IN 46307
219-662-3800 (v)
219-662-3877 (f)
www.customspeechusa.com
 info@customspeechusa.com
We teach computers how to listen®

 Use our Online Information Request Form

You are receiving this copy of the Custom Speech USA, Inc. newsletter because you have expressed interest in our products and services.  If you have been sent a newsletter in error, or would like to be removed from the list, please click here. 

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.  Click here for Online Privacy Policy. 

Dragon and NaturallySpeaking® are licensed trademarks of Nuance® (Nuance Communications, Inc.)