Let There Be Speech!™

Affordable Server-Based Speech Recognition . . .


"In the hospital Informatics department, we are developing a wireless mobile communication system to support physician voice queries of the relevant research literature in their specialty.
After a web search for a low-cost solution and discussion with Custom Speech, we selected their SpeechServers running with Dragon NaturallySpeaking, medical version. During the
setup process their technical support was friendly and knowledgeable. The system is up and running and we recommend it to anyone looking for cost-effective software for server-based
speech recognition."
Robert Duffy, Senior Programmer, BioMedical Informatics, Columbia University, New York, NY

Someone once said a long time ago that this country needs
a good 5-cent cigar. Something similar can be said of server-based speech recognition


According to a 2006 survey, many physicians preferred server-based, back-end speech recognition because it did not require any significant change in their dictation habits.  Start-up costs for many speech recognition applications were estimated at $900 out of box, or less than $3,000 in the first year including training, support, and maintenance, to $4,000 to $10,000 per user, with lower costs in subsequent years. E. S. Roop, "Speaking of Savings: Can Speech Recognition Deliver?," For the Record (May 15, 2006)  Given the inevitable errors and required editing of speech recognition text, should you pay up to $10,000 per speaker for server-based speech recognition?    

SpeechServers™ (U.S. and foreign patents, patents pending) supports server-based speech recognition for SAPI 5.x (e.g., SweetSpeech, Windows Vista), Dragon, and IBM speech engines.¹  The SpeechServers™ software uses off-the-shelf or runtime licenses for Dragon and IBM speech engines,² and an earlier version of the Microsoft speech engine is available FREE.   More recent versions come with the operating system.   The "servers" can be integrated with the Command! workflow manager for file and speech user management, document retrieval, time tracking, web services, and telephone dictation.   Licensing is based upon usage.  The system is designed for use with the  SpeechMax™ HTML text editor.  Buy one Dragon license, and get a FREE Dragon license for  transcriptionist.³    SpeechServers™ is available with SpeechProfessional™ combo package.  SpeechServers™ SAPI 5.x is included in same install kit as SweetSpeech™, but is separately licensed.

Key Features:

  • Features and languages supported dependent upon underlying speech engine

  • User profiles may be shared by multiple servers or for real-time speech recognition
  • Use of speech engines subject to any applicable per-speaker license fees
  • Supports Windows 2000/XP/2003/Vista
  • Compatible Speech Recognition Engines
  • SAPI 5.x  (e.g., Microsoft Vista, SweetSpeech)
  • Dragon v. 8.1+ (Professional, Medical, Legal)*
  • IBM v. 8+ (Professional)*
  • Use with only most recent versions of third-party software supported
  • Includes separate license for Command!
  • Command! required to configure SpeechServers
  • Licensing
  • Based upon usage like phone card
  • 6M word limit
  • Purchase "unlimited words" license add-on for $99 per each speech engine

*System may run with Dragon Professional, Medical, or Legal v. 8.10.000.,285 or higher, but only version 9.x is supported.  System may run with IBM Professional v.8.x or higher, but only  IBM USB Pro 10.x is supported.  

SweetSpeech/SpeechServers SAPI 5.x ("SAPI 5.x"):

  • Text output (TransWaveX)

  • Custom Speech USA .ses session file (SaveSession)

  • Untranscribed .ses session file (SpeechSplitter)

  • SpeechSplitter segments speech and nonspeech audio, including songs and music

  • Microsoft SAPI 5.x speech engine available FREE

  • Requires SAPI 5.1 or higher

  • Use with patents pending SweetSpeech

  • SpeechServers SAPI 5.x in same install kit as SweetSpeech™, but separately licensed

SpeechServers™ ("Dragon")

  • For Dragon v. 8.1 or higher, however only 9.x supported
  • Text output (TransWaveX™)
  • Dragon .dra session file output (SaveSession)
  • Custom Speech USA .ses session file output (SaveSession)
  • Transcribe with off-the-shelf Dragon Professional or runtime
  • Open .dra in SpeechMax™ or Dragon interface
  • Open .ses in SpeechMax™ only
  • Custom Dragon Server Edition development available

SpeechServers™ Dragon v. 5/6 and IBM v. 8+ ("Dragon/IBM")

  • Text output (TransWaveX™)
  • Automated corrective (SpeechTrainer™)
  • Custom Speech USA .ses session file (SaveSession)
  • Create speech user from audio and text (Complete Profile)
  • Standalone version available
  • Compatible with IBM Professional v. 8.0 or higher, however only 10.x supported
  • Use with only most recent versions of third-party software supported
  • SpeechServers for Dragon v. 5/6  no longer supported
SpeechServers™ Dragon SAPI 5.x Dragon/IBM
TransWaveX™
SpeechTrainer™
SaveSession™
Complete Profile™
SpeechSplitter™
Standalone (N/A)
Dragon v. 8.1+
Dragon v. 5/6
IBM v. 8+

Related utilities:

  • SpeechUsers speech user management of speech user files

  • SpeechSettings™ job configuration

  • Included with SpeechServers

Dialog when running a Dragon v. 6 speech engine and four services:



See for yourself--click here for video demos

_____________________________________________________

¹ Server-based and real-time speech recognition is available for Dragon Professional, Medical, and Legal 9, IBM ViaVoice Professional 10, and SAPI 5.x speech recognition, e.g., this company's speaker-specific SweetSpeech and speech recognition included FREE with Windows Vista OS.  System also supports SAPI 5.x text to speech, including Microsoft, AT&T Natural Voices, and NeoSpeech VoiceText voice fonts.  System may run with Dragon Professional, Medical, or Legal v. 8.10.000.285 or higher, but only version 9.x is supported.  System may run with IBM Professional v.8.x or higher, but only  IBM USB Pro 10.x is supported.   See other products for pricing and more information.

²  Dragon and IBM speech recognition dictation-only runtimes may be purchased from Custom Speech USA only for use with the company's products.  Electronic Help manual is included with the runtime, but not voice commands, printed manual, or headset microphone.  AT&T Natural Voices and NeoSpeech VoiceText voice fonts runtimes are available for use only with the company's products.   See other products for pricing and more information.
 

³  Dragon software must be provided by speaker and used solely to edit his/her documents.
 

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