SpeakScape:
Voice
notes and conversion archiving and retrieval
SPKS is an automated voice notes conversion infrastructure that requires a recording device that saves
voice notes (WAV format). The user may also use a laptop or desktop PC, and a hand-held PDA recorder with wireless communication. The rest is automated. Wav files are either fed directly to the
system for conversion or uploaded from the recording device for batch conversion. Documents are converted and saved in the index of the search
system for searching exact content of recorded voice notes. All converted
documents are created in a quadruplet structure (polytuplet) so that during search, a user may click the original converted
recorded voice file, snippets (smaller pieces) of the file, a text file
of the content and an XML file that binds all the various pieces
of the converted voice note together. Features are described in detail below.
Uses
Anywhere dictation or recorded messages can be taken, they are converted to a
text format and stored for quick access using folders or pattern
profiling search. SpeakScape is ideal for legal and medical dictations, sales and real estate reports, general dictation and writing, security observations, loan officer anti-fraud observations, police reports, etc.
SPKS can be used for batch archive conversion or as a real-time feed and search (profiling) system. Case files are paired to recorded observations or notes, for example. Simply
upload voice recordings (notes) to the system or record
directly and the system will automatically convert the voice
notes for browsing and viewing or listening to results. Or use search
to find snippets or full recordings with the content you are
looking for. The transcriptionist can use the text files of each
snippet or total converted file and begin editing and formatting
processes.
SpeakScape |
# of
Users |
Error
Correction |
Auto-Sort
& Collate |
User
Mgmt |
SPKS-1 |
1 |
|
x |
x |
SPKS-2 |
1 |
x |
x |
x |
SPKS-3 |
5 |
x |
x |
x |
SPKS-4 |
10 |
x |
x |
x |
SPKS-Enterprise |
100 |
x |
x |
x |
Embedded in the ScapeShape framework for rapid deployment
Packaged with search, forum discussions, RSS feeds, announcements, file sharing, and controls
for an extensive Intranet access system for content of your
voice notes, paperless converted notes and existing network and computer
documents.
Error correction
Error correction is performed automatically in a synonym mapping process where error words under speech recognition are mapped to correctly spelled words. Consequently documents are found during search even when a search term is stored incorrectly as it has been mapped to the correctly spelled equivalent (spell-synonym).
Large wav file splitter
Wav files are split into snippets to facilitate search, find, and listening speed. The polytuplet presentation
in search results includes the original wav file, its multiple
segmented snippets, and a text converted PDF for analysis. This way, a user does not need to listen to long recordings to find results, rather
he can review snippets for reconfirmation before using large converted files.
Handwriting conversion option
User recorded Wav files and Tiff images of handwritten documents are integrated to facilitate searching these documents. Search results include the original handwritten Tiff image, a recorded transcription snippet file, and a text converted PDF polytuplet result. The user simply scans an image, reads as much of the handwritten document into his headset and the system integrates the handwritten Tiff and the voice transcription for search purposes. Pattern profiling search and conversion technologies render hand written documents as searchable assets. Note: any typed (form) content on a handwritten document is also indexed to facilitate search.
Naturally the transcription process can be as brief or long
as user needs for searchable mapping granularity. The user
interface also allows for assigning a saved-snippet to all
scanned documents within a project. Thus, snippet integration
can be auto-integrated to scanned documents thereby increasing
productivity of a batch conversion project.
Automated sort and collating folders
All wav documents are saved in folders based on the content of the document. This is done using the PS DocMap process for mapping
content and creating a folder system based on the mapped
clusters. The administrator only need create a table of reference for the system to
collate documents correctly based on content. For example, in a clinic application, doctor names would be root folders while
patients' folders would be sub-folders. Clinical assessments, billings, treatments, Rx, notes and instruction folders would propagate below the patient folder; this is all done automatically based on
the document content reference table.
This provides for automated scan-and-convert archiving with collation and folder
structuring (all this wrapped in an extensive search
infrastructure). Regardless of the size of the archival or
escrow project the system can save, index and search vast amounts of documents thereby providing an industrial
class alternative to DVD (CD) storage and retrieval solutions.
Containers
All recorded Wav files (voice notes) and converted TXT files are stored in SpeakScape folders under ScapeShape
folders and containers for each user. Containers are capable of containing folders for every user or designation determined by your administrators.
Search
Pattern profiling search is through the 2x2 Search Pad providing excellent search and find capability
for large paperless and voice notes archives. SPKS leverages valuable voice
notes and interviews into valuable information and intelligence
assets.
|