Unlock the Power of Your Voice™
Speaker-specific speech recognition can be more accurate . . .
--Create accurate speech user
profiles from day-to-day dictation
--Profile created "behind the scenes," no speaker enrollment required
--Unlimited speaker license available
The primary emphasis in
speech recognition, text to speech, and other speech and language processing in
research has been has been working with data from large numbers of speakers or
authors. In speech recognition, the focus has been speaker
independent models, or speaker adaptive models that use speaker independent data
and are adapted to a speaker's voice through enrollment and corrective training. Some
have proposed a focus on the individual speaker and using "massive amounts of
speaker-specific training data recorded in one’s daily life. We call this
Massively Speaker-Specific Recognition (MSSR) . . . . Initial results show that
by changing the focus to MSSR, word error rates can drop very significantly.
In comparison with speaker-adaptive speech recognition system, MSSR also
performs better since model parameters can be tuned to be suitable to one
particular individual."
Y. Shi & E. Chang, "Studies in Massively Speaker-Specific Speech
Recognition" (IEEE 2004)
Here is your toolkit
for speaker-specific processing!
Editions:
Full -- includes full speech and language toolkit
Audio Segmentation (SpeechSplitter™) -- audio/voice/music segmentation
Lexicon/Language Model -- (WordPronounce™/WordContext™)
Patents
pending Windows-based SweetSpeech™ toolkit includes Model Builder to create
acoustic model, language model, and lexicon for speaker-specific speech
recognition. The SAPI 5.x-compatible toolkit further includes phonetic
pronunciation generator and regular expressions application for forward and
reverse formatting text. Using day-to-day dictation audio and
transcription, create one or more speech user profiles from audio and text data
that was formerly discarded as a useless byproduct of the dictation process.
The system supports Unicode and end-user adjustment of processing parameters.
SpeechServers™ SAPI 5.x is included in same install kit as SweetSpeech™, but
is separately licensed.
The speech engine may be used with the
SpeechServers™ user and file management for
back-end, server-based
speech recognition. Plugins for the
SpeechMax™ HTML session file editor
support real-time interactive speech recognition with the system and local,
client-based transcription of an audio file. The session file editor
may also be used in the preautomation, training phase to generate training data
for the speech user profile, and in the automation phase, to edit server-based
or real-time speech recognition. Tools for audio segmentation, language
model, and lexicon may be ordered separately. Use the the toolkit with
SpeechProfessional™
software suite and purchase at a discount as an
EnterpriseSpeech™
add-on.
SweetSpeech™ toolkit is a "best-buy" value.
Compare with conventional off-the shelf large vocabulary, continuous speech
recognition systems.
|
