USB DICTATION DEVICE
BACKGROUND OF THE INVENTION
Dictation devices have been in common use in many fields in which it is inconvenient or
undesirable to make handwritten or typewritten notes. One of the fields in which dictation
devices have long been prevalent is the medical profession, and particularly so among
radiologists, who often dictate their findings and diagnoses while referring to a photographic
print of radiological data such as X-rays. It is common in these fields for a user of a dictation
device to make voice recordings and provide these recordings to transcriptionists, who transcribe
the recordings in order to generate written transcripts of the recording for the dictator's review or
for record keeping purposes.
More recently, dictation technology has developed significantly and includes such tools
as speech recognition software to eliminate some of the need for transcriptionists to transcribe
recordings. However, a recognized and pervasive problem in the art is that speech recognition
requires high quality audio input. Low quality audio input decreases the effectiveness of speech
recognition algorithms, and frequently prevents them from functioning at all. It is therefore
desirable to provide a microphone with superior audio quality for connecting to a computer for
implementing speech recognition.
In some applications of dictation devices and speech recognition tools, a dictator uses a
dictation device and speech recognition to complete blank text fields in a form. For example,
many medical practices have specific forms wherein there is a printed query or prompt, followed
by a blank text field into which a practitioner provides the requested text information regarding a
particular patient. One way in which different fields are selected and activated to receive text
from a voice recognition interface is by use of a pointing device, such as a mouse. However, this
manner of selecting various fields for text input is unduly awkward as typically a user would
prefer to use the same hand to manipulate the microphone as the mouse. Thus, it is desirable to
provide an ergonomically convenient way to navigate through forms containing fields in which
text is entered through a dictation device and a speech recognition interface.
Technological advances have led to newer and faster types of interfaces between
peripheral devices and computers. Dictation microphones for connecting to computers running
speech recognition software have been known in the art. However, these microphones connect to
computers through one or more serial ports, and often require other connections as well,
including speaker in, audio in, speaker out, audio out, power, RS232, and game port connections.
These multiple connections make connecting the microphone to a computer a time consuming
and complex process. Furthermore, many dictation microphones require connection to a sound
card, which many laptop and some desktop computers lack. Therefore, a substantial number of
computers are unable to connect to these sorts of dictation microphones. However, the
development of the USB and USB2 standards ("Universal Serial Bus" and "USB2" are both
hereafter simply referred to as "USB") have brought significant increases in speed and aided in
uniform c ompatibility b etween p eripheral d evices and c omputers. It is desirable to provide a
dictation microphone that can be connected simply and easily to a computer through a USB
connection, through which two way communication between the microphone and the computer is
established, and from which the microphone can draw substantially all of its power requirements.
In some applications a barcode identifier is used to identify the subject about which the
user of a dictation device is dictating. For example, in the field of radiology, it is common for a
radiologist to use a scanner to scan a barcode on a radiograph, such as an X-ray, which identifies the patient and/or the X-ray. The radiologist then records his findings and diagnoses into the
dictation device with the assurance that this recording will be associated with the correct patient
and/or X-ray. It is therefore desirable to provide a dictation device that may include a barcode or
other such scanning ability.
A difficulty arises, however, because scanning devices generally, and in particular laser scanning devices, generate electromagnetic fields that can interfere with microphone circuitry and degrade audio signals, thus making speech recognition of those signals less accurate or impossible. Therefore, there is a need for circuitry that can overcome the effects of interference between the scanning and microphone elements in a dictation device having a scanner. It is desirable to have an integrated scanner/dictation device that can interface with a computer
through a USB connection, and which can draw all of its power requirements from the USB port. However, the USB standard limits the amount of current that can be drawn to 500mA, which may be insufficient to drive both the scanner and the dictation device.
OBJECTS OF THE INVENTION
In light of the above identified deficiencies of the prior art, an object of the present
invention is to provide to provide a microphone with superior audio quality for connecting to a computer for implementing speech recognition. It is another object of the present invention provide an ergonomically convenient way to
navigate through forms containing fields in which text is entered through a dictation device and a speech recognition interface.
Yet another object of the present invention is to provide a dictation microphone that can be connected simply and easily to a computer through a USB connection, through which two way communication between the microphone and the computer is established, and from which the microphone can draw substantially all of its power requirements.
Still another object of the present invention is to provide to provide a dictation device that may include a barcode or other such scanning ability.
Another object of the present invention to provide an integrated scanner/dictation device that can interface with a computer through a USB connection which can draw all of its power requirements from the USB port.
SUMMARY OF THE INVENTION
The invention is a handheld device for receiving audio input in the form of speech, the
audio input recorded in a recording medium and processed by a speech recognition engine,
thereby generating text. The handheld device may comprise an omni-directional microphone
element disposed at a distal end of the device for receiving the audio input in the form of speech
and for generating an analog signal therefrom. The handheld device may further comprise a wind
screen selected according to a predetermined wind noise sensitivity factor. The wind screen may
be acoustically transparent and attenuate wind (air flow) noise for example with hard consonants,
that create too great an analog signal and would otherwise disrupt the speech recognition. The
handheld device may further comprise an electric circuit for receiving the audio input in the form
of speech and for converting the analog signals to digital signals. The electric circuit may have a
gain control for providing a signal level that is suitable for speech recognition.
The handheld device is preferably used in conjunction with a software program that
utilizes a speech recognition engine to fill in blank portions of forms, thus allowing a user to
input text in a blank portion of a form by simply speaking the words to be entered in that portion.
The handheld device may include a set of input buttons ergonomically positioned on the top
surface of the handheld device. The input buttons may include one or more buttons for
selectively navigating through predetermined sections of a form, a button for selecting a
predetermined section of a form, and a button for initiating recording.
The navigation buttons may include a button for advancing to the next predetermined
section of a form and a button for going back to a previous predetermined section of a form. The
buttons may further include a select button for selecting the active element in a form. The
navigation buttons may be used to navigate through various elements in a form by sequentially
activating those elements, then when a particular desired element is activated, the select button may be pressed to select the active element. Once an active element has been selected, text may
be inserted into that element. The input buttons may include a button for initiating dictation.
Preferably, text generated by the speech recognition engine is not immediately displayed as it is recognized because it has been found that for many users, this would be too distracting. Thus when dictation has been initiated by the user by pressing the dictation button, the text generated by the speech recognition engine is held in a buffer until the user presses an "insert text" button. When the user presses the "insert text" button, the contents of the buffer are displayed in the selected element of the form. In an alternative embodiment, the text generated by the speech recognition engine may be displayed immediately after the speech has been recognized and while the user is dictating.
Other functions of the buttons may include selecting a portion of text, playing a portion of an audio file, a stop button, a button for reviewing back through an audio file to a previously recorded portion of the file, a button for advancing through an audio file, and a button for playing a portion of an audio file through an integrated speaker on the handheld device. In one embodiment, the button for playing a portion of an audio file is also a stop button, such that if any function of the microphone is active, pressing the play/stop button will stop any active function. If no function of the microphone is active, pressing the play/stop button will play a portion o f a n a udio file. T he h andheld d evice m ay a lso i nclude a b utton f or d ictation which when pressed activates the microphone element to receive audio input in the form of speech. When this element is active, the audio input is digitized and transmitted to the speech recognition engine, which then translates the speech into text.
The handheld device may also include one or more buttons with programmable functions. These buttons may be assigned a function that may generally be quite complex by way of user
recorded or prerecorded macros. In a preferred embodiment, the handheld device includes a "signature" button for electronically signing a form after it has been filled out. Actuation of this
button designates the current document signed, the current document is saved and further editing is blocked. For example, if the handheld device is being used by a medical doctor, the doctor may complete a form and review it for accuracy, then press the "signature" button, which adds a marker to the document indicating that it has been completed and reviewed, and that the doctor acknowledges that the information contained in the document is true and complete. After the "signature" button is pressed, the document may be saved and marked as read-only to prevent further editing of the document.
Another button that would be particularly useful for medical professionals is a "coding"
button. When the "coding" button is pressed, a dialog box may appear containing diagnosis numbers. Particular maladies are typically assigned unique codes. A doctor can press the "coding" button and view a menu or other dialog box and select a numerical code corresponding to the diagnosis. This code may then be appended to the document. The numerical codes simplify billing procedures because most insurance carriers rely on these codes rather than a textual description of the maladies treated in order to determine how much to pay for particular procedures. The codes may also provide a database to ease the process of searching, for example, for patients with particular maladies. Further details on coding operations can be found in co-pending U.S. Provisional P atent Application S erial N o. 60/436,456, filed D ecember 27, 2002, incorporated herein by reference in its entirety.
Preferably, the buttons are ergonomically positioned to allow one-handed navigation
through the predetermined sections of a form, selection of a predetermined section of a form, initiation of recording, and insertion of text. The ergonomic positioning of the buttons preferably
allows one-handed operation of every function available on the microphone. The handheld device may include a pointing device on its top surface that may be
actuated by the user's thumb. Preferably, the pointing device is a microjoystick or thumbstick, although it may also be a trackball or touchpad, or any other suitable pointing device. The pointing device has associated with it at least one button. Preferably, there is a button on the top of the handheld device, just below the pointing device. Most preferably, there is a button on the bottom of the handheld device, beneath where a user's index finger would naturally rest, such that advanced mouse maneuvers such as drag-and-drop may be performed by holding down the
button with the index finger while moving the pointing device with the thumb.
In one aspect, the invention is a handheld USB device for receiving audio input in the form of speech. The audio input may be recorded in a recording medium and processed by a speech recognition engine, thereby generating text. The handheld device may include a USB hub for receiving and transmitting signals through a USB interface to a USB root hub in a computer. It may further include a power switch for switching power drawn from the USB interface between a first USB port on the USB hub and a second USB port on the USB hub. It may further include a pointing device connected to the second USB port on the USB hub. It may further include a USB streaming controller connected to the first USB port on the USB hub. The USB sfreaming controller may receive digital audio signals from an audio codec, which converts analog audio signals from a microphone element on the handheld device into the digital audio
signals.
Preferably, the handheld USB device includes a p reamplifier b etween the audio codec
and the microphone element for amplifying the analog audio signals from the microphone
element before they are digitized by the audio codec into digital audio signals. The handheld
USB device also preferably includes a speaker. Digital audio signals from the USB streaming
controller may be converted into analog signals by the audio codec. These analog signals may
then be amplified by an amplifier, and the amplified analog audio signals then used to drive the
speaker.
The handheld USB device may include buttons connected to switches, such as USB
human interface devices. The buttons and switches may be arranged such that when a button is
pressed, its corresponding switch is closed. Alternatively, when a button is pressed, its
corresponding switch may be opened, if the default position of the switch is closed. The USB
streaming controller may be used to detect whether a switch is opened or closed, and thereby may
send a signal through a USB port on the USB hub, which then transmits the signal through a
USB interface to a USB root hub on a host computer. The USB streaming controller may also be
connected to one or more LEDs, which the USB streaming controller may light in response to a
signal indicating that a button has been pressed. Preferably, the USB streaming controller lights
an LED when the button corresponding to the "record" function of the handheld USB device is
pressed.
In one embodiment, the handheld USB device may include a data storage device.
Preferably, the data storage device is a memory stick or a SmartMedia™ card. In other
embodiments, the data storage device may be any data storage means known to those in the art.
In another embodiment, the handheld USB device may include a fingerprint security device. Preferably, the fingerprint security device will lock out any unauthorized user, thus preventing
unauthorized users to use the handheld USB device to create or alter medical records. The
handheld USB device may further include a barcode scanner for scanning barcodes, for example,
on medical records to identify the patient, or on pharmaceutical packaging to ensure patients
receive the correct medicines.
The above advantages and features are of representative embodiments only, and are
presented only to assist in understanding the invention. It should be understood that they are not
to be considered limitations on the invention as defined by the claims, or limitations on
equivalents to the claims. Additional features and advantages of the invention will become
apparent from the drawings, the following description, and the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
While the specification concludes with claims particularly pointing out and distinctly
claiming the present invention, it is believed the same will be better understood from the
following description taken in conjunction with the accompanying drawings, which illustrate, in
a non-limiting fashion, the best mode presently contemplated for carrying out the present
invention, and in which like reference numerals designate like parts throughout the figures,
wherein:
FIG. 1 is a top view of one embodiment of the dictation device of the invention.
FIG. 2 is a bottom view of one embodiment of the dictation device of the invention.
FIG. 3 is a block diagram of one embodiment of the dictation device of the invention
without an integrated scanner.
FIG. 4 is a block diagram showing the data flow among the various components of one
embodiment of the dictation device of the invention.
FIG. 5 is a block diagram of one embodiment of the dictation device of the invention with
an integrated scanner.
FIG. 6 is a block diagram of one embodiment of the dictation device of the invention with
fingerprint security and memory stick features.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The present invention encompasses dictation microphones and dictation microphones
with integrated scanning units. The invention includes a novel ergonomic layout for facilitating
use of the dictation microphones with application software implementing voice recognition
algorithms to reduce voice data to text and to place the text in appropriate blank fields in a form.
The novel ergonomic layout further includes a pointing device with buttons positioned to
facilitate pointing device functionality, such as drag-and-drop. The invention solves a number of
problems associated with prior art methods of interfacing dictation devices with computers and
provides a convenient USB interface. The invention further solves power limitation problems
associated with the integration of dictation devices with scanning devices using a USB interface.
It is known in the art that hard consonant sounds, like -d, -t, -p, -b, -x, and -z have higher
instantaneous energies, that is they create higher wind speed than other sounds, and can reduce a
microphone's fidelity, and thus cause speech recognition rates to drop. This effect is deemed
"Wind Noise Sensitivity" (WNS). The inventors have developed a test for determining whether
a microphone and windscreen combination is adequately resistant to these wind noises. For an
electret microphone with an output sensitivity of -44 dB (1 KHz, 0 dB = 1 V/pA), the WNS
should be less than -60 dB at a wind speed of 1 meter/sec in order to provide an audio fidelity
adequate for voice recognition. It is anticipated, however, that increasingly sophisticated voice
recognition algorithms may overcome the limitations of noise introduced by hard consonant
sounds, and that a greater WNS would be adequate. One aspect of this invention is a method for
determining whether a particular microphone and windscreen combination is adequate for use
with a voice recognition algorithm.
FIG. 1 shows a top view of an embodiment of the microphone of the invention 101. The
microphone element 102 may be protected by a windscreen consisting of a layer of foam padding
material, preferably polyurethane foam, with a thickness of 0.5 to 1.6 cm and a porosity of
between 60 and 90 ppi (pores per inch). Preferably, the windscreen is a polyurethane foam with
a porosity between about 100 and 130 ppi and a thickness of between 0.6 cm and 1 cm, most
preferably 0.9 cm. Most preferably, the windscreen minimizes the wind noise sensitivity of the
microphone yet has a flat and high transmission profile across the audible frequency range
(approximately 20 Hz to 20 kHz). Preferably, the windscreen is affixed to the microphone by
being placed such that its sides are pinched between the microphone casing and the microphone
element itself in such a manner as to make it flush with the surface of the microphone. This
design prevents shear forces from detaching the windscreen. Preferably, glue is not used to affix
the windscreen to the microphone because the glue could dislodge and get into the microphone
element, thus degrading the analog signal from the microphone and making speech recognition
difficult or impossible.
The microphone 102 is preferably a close-talking microphone. Preferably, the
microphone element is adapted to receive audio input from a speaker whose mouth is between
0.5 and several inches from the microphone element. Preferably, the microphone element can be
as much as 45 degrees of mouth axis. The large variation in frequency response of unidirectional
and noise-canceling microphones under those conditions make these microphones ill-suited for
this application because of the resulting decrease in the accuracy of speech recognition.
However, these microphones maybe used in this application if users are cautious about how they
hold the microphone, or if the microphone is mounted in a fixed relation to the user's mouth.
The microphone element is preferably an omni-directional microphone with a frequency response
that is substantially flat over a range of angles and distances.
A plurality of buttons 104-111 allow the user to control the dictation functions. A "next"
button 104 allows the user to advance to the next field in a form. A "previous" button 1 05
allows the user to go back to the previous field in a form. A "select" button 106 allows the user
to select the current field in a form. An "insert text" button 107 allows the user to insert text at
the current position of the cursor that has been stored in a buffer during dictation. A "dictate"
button 108 begins the recording process, allowing the user's speech to be recorded. An LED
indicator 103 is lit when the microphone is recording. A "fast forward" button 109 allows the
user to skip through previously recorded speech to search for a particular portion. A "review"
button 110 allows the user to skip backward through previously recorded speech to search for a
particular portion. A combination "stop/play" toggle button either allows the user to play
previously recorded speech, or to stop playback of previously recorded speech or any other active
function of the microphone.
A thumbstick pointing device 112 is included to allow the user to navigate through a form
document or for any other reason that a mouse pointing device is normally user for. The
thumbstick pointing device also includes a mouse button 113. In a preferred embodiment, the
pointing device is a force sensing resistor micro joystick pointing device. In alternative
embodiments, the thumbstick pointing device can be any other suitable pointing device, such as a
trackball.
Two additional buttons (not shown) are optionally present. Generally, these two buttons
can be programmable to meet individual users' needs. In a preferred embodiment, actuation of
one of the additional buttons executes a software routine to provide an electronic signature for
signing forms after they have been filled out using the microphone and voice recognition
technology. In this embodiment, actuation of the signature button adds a "signed" notation to the
form after it has been filled out, saves the completed form, and marks it read-only. In this
embodiment, actuation of the other additional button executes a software routine to provide a
menu of codes corresponding to a malady which a doctor has diagnosed in a patient. The doctor
selects the code corresponding to her diagnosis, and the code may be associated with the form or
forwarded to the patient's insurer for billing purposes. Either of these two additional buttons
may be programmed to have an arbitrary function. For example, one of the buttons may be
programmed to automatically generate and send an email message containing a recently recorded
digital audio file recently recorded on the handheld dictation device to a predetermined recipient,
as disclosed in co-pending U.S. Patent Application S erial No. 09/099,501, entitled "Dictation
System Employing Computer-to-Computer Transmission of Voice Files Controlled by Hand
Microphone," filed June 8, 1998, and incorporated herein by reference.
FIG. 2 shows a bottom view of an embodiment of the microphone of the invention 201.
A speaker 202 may be included for playing back previously recorded speech or other wav files.
A button 203 may be used in conjunction with the thumbstick pointing device 112 and is
preferably corresponds to a left mouse button. This allows the user to manipulate the thumbstick
pointing device 112 while simultaneously holding down the button 203, thus allowing for mouse
operations such as drag-and-drop. At the base of the microphone 204 may be a slot for
removable memory. Also included may be a space 205 for an integrated barcode scanner.
FIG. 3 shows a block diagram of an embodiment of the microphone of the invention. In
this embodiment, a host PC 301 having a USB root hub 302 may be connected to the microphone
303 through a USB connector 305. The USB connector connects to a USB hub controller 304 in
the microphone 303. The USB hub controller 304 in the microphone 303 can interpret the
various signals coming from the elements on board the microphone 303 in order to send those
signals in a meaningful way to the host PC 301. The USB hub controller 304 thus allows
multiple elements in the microphone 303 to send signals to the host PC without the signals
becoming intractably entangled. In a preferred embodiment, the USB hub controller 304 is
comprised of a Texas Instruments™ TUSB2036 2- or 3-port USB hub chip. A power supply
voltage regulator 307 may convert +5V routed from the USB connection to the +3.3V needed to
power the USB hub controller chip 304. In a preferred embodiment, the USB hub controller 304
is powered by a Texas Instruments™ TPS78833 power supply voltage regulator 307.
A power switch 306 may provide power management for the downstream devices in
order to comply with USB power management requirements. Preferably, the power switch 306 is
a Texas Instruments™ TPS2044 chip. Both output ports of power switch 306 are tied together,
and the +5V output is directed to two power supply voltage regulators 308 and 309. In the event
of an overcurrent, the power switch 306 can switch off the power to the power supply voltage
regulators 308 and 309. When the power switch 306 is active, thus sending +5V to the power
supply voltage regulators 308 and 309, the power supply voltage regulators may convert +5V
from the USB connection to the +3.3V needed to power the downstream elements. Regulators
308 and 309 may independently provide +3.3V to the digital components of the handheld device
(regulator 308) and the analog components of the handheld device (regulator 309). The
independence of the two regulators allows a highly uniform voltage source to be provided to the
analog elements (the microphone 314, speaker 315, and amplifiers 319 and 320), regardless of
the power requirements of the digital components (the EEPROM 311, the audio CODEC 318,
and the streaming controller 312).
Under the USB standard, only 500mA of current can be drawn from the USB connection.
Preferably, high-power functions will draw less than 100mA at power up. Thus, the power
switch 306 serves the additional function of shutting down the downstream devices if they
attempt to draw more than this maximum amount of current. Note that the power to the USB
hub controller chip is preferably powered by an independent power supply voltage regulator 307
that is not switched by the power switch 306. In a preferred embodiment, the USB streaming
controller 312 is attached to a current sensing device or other sensor 321. When the sensor 321
senses an overcurrent or the activation of an element with excessive current requirements, the
USB streaming controller 312 can send a signal 322 to the speaker amplifier 320 that mutes the
speaker 315, thus reducing the current requirements of the handheld device. In the embodiment
of the invention shown in Fig. 3, there is no barcode scanner, and thus it is not expected that a
current in excess of 500mA would ever be required. Thus the sensor element 321 is not strictly
necessary in this embodiment. However, the mute signal 322 may still be sent by the host
computer 301 when the recording mode is activated in order to prevent sounds from the speaker
315 to be picked up by the microphone 314 during dictation.
In a preferred embodiment, the dictation microphone 303 has an on-board pointing
device, preferably a microjoystick pointing device 310. The pointing device 310 is connected to
one of the ports of the USB hub controller 304. Microjoysticks that may be used with the present
invention can be obtained from Interlink Electronics™.
The microphone 303 further includes an electrically erasable programmable read only
memory (EEPROM) 311 for storing instructions including CODECs and input and output data
bit rates for the USB streaming controller 312. The EEPROM 311 is connected to the L2C port
of the USB streaming controller 312. The EEPROM can be programmed or reprogrammed by
way of signals from the USB root hub 302 through the USB connection 305 to the USB hub 304
and the USB streaming controller 312 using for example, a device firmware upgrade utility
which is compliant with USB Device Class Specification for DFU 1.0. Preferably, the EEPROM
is a Microchip Technology Inc.™ 24LC64 chip.
A USB s treaming c ontroller 3 12 receives the input from several switches 313 and the
microphone element 314, and provides the output to the on-board speaker 315 and two light-
emitting diodes (LEDs) 316 and 317. Each of the LEDs may be lit under certain specific
circumstances, for example, when the handheld device is plugged in one LED may be lit to
indicate the device is ready to be used, and the other LET may be lit during the recording or
dictating process. The switches 313 correspond to the buttons 104-111 described with reference
to FIG. 1, as well as the two programmable buttons, such that actuation of any of those buttons
results in the closure of the corresponding switch. In a preferred embodiment, the USB
streaming controller is a Texas Instruments™ TAS1020A chip.
Also connected to the USB streaming controller 312 is an audio coder/decoder (CODEC)
for converting analog signals from the microphone 314 into digital signals that can be transmitted
to the PC host 301 or stored in an optional on-board memory (not shown). Between the
microphone 314 and the audio CODEC 318 is a fixed gain front end amplifier 319 for amplifying
the analog signal from the microphone to take advantage of a greater dynamic range before the
signal is digitized. The audio CODEC 318 also serves the function of converting digital signals
from the PC host 301 or stored in an optional on-board memory (not shown) into analog signals,
which can be played on the on-board speaker 315. Between the audio CODEC 3 18 and the
speaker 315 is an amplifier 320 for amplifying the analog signal to produce an adequate volume at the speaker 315. In a preferred embodiment, the audio CODEC 318 is a Wolfson™ WM9707
chip, the fixed gain front end amplifier 319 is a National Semiconductor™ LMV110 chip, and
the speaker amplifier is a 0.35W mono audio power amplifier comprising a Texas Instruments™
TPA301 chip. Also in this preferred embodiment, the microphone 314 is an electric condenser
microphone such as Panasonic's™ WM-52M and the speaker 315 is a Panasonic™ EAS2P104H
micro speaker.
FIG. 4 is a block diagram showing the data flow through three conceptual "layers" — a
function layer 410, a USB device layer 460, and a USB bus interface layer 490. Solid arrows
represent d ata flow and hollow arrows represent logical communications flow, or instructions
provided from one component to another. The client software 400 can be any voice recognition
software, especially voice recognition software as part of a larger application that allows forms
with blank fields to be retrieved and where the blank fields can be filled in using voice
recognition. The USB system software 450 and USB host controller 480 can be any suitable
USB system and host controller software, for example those provided with typical operating
systems having USB capability, such as Windows™ or Linux. The USB hub 490 is internal to
the dictation device, and corresponds to element 304 in FIG. 3 for separate control of a USB
Streaming Controller and a pointing device controller. The USB device layer includes a USB
streaming controller 470 and the USB pointing device controller 465, each with a data
connection to the USB hub 490.
The function layer includes an audio function 430, for receiving analog audio data from
the microphone physical device and converting the analog data to digital for transmission
through the USB streaming controller 470 to the USB hub 490, through the USB host controller
480, through the USB system software 450, to the client software 400, where voice recognition
algorithms converts the audio data to text, and places that text in the appropriate place, for
example in a form document. Within the audio function 430 is the capability to output to the
speaker physical device. The client software sends digitally encoded audio data through the USB
system software 450, the USB host controller 480, the USB hub 490, the USB streaming
controller 470, to the audio function, which converts the digital data to analog, which is then
amplified, and the amplified signal drives the speaker physical device.
The function layer also includes several input/output (I/O) functions 420, for receiving
input in the form of an identity of an actuated button. This data is likewise transmitted through
the USB streaming controller 470, the USB hub 490, the USB host controller 480, the USB
system software 450, to the client software 400, which recognizes the actuated button as
corresponding to a command to be performed in the client software application.
The function layer also includes a pointing function 415, which receives data from the
pointing physical device and buttons. The pointing data is routed through the USB micro
joystick controller 465, through the USB hub 490, through the USB host controller 480, through
the USB system software 450, to the client software 400, where the pointing data is used, for
example, to navigate through a form having blank fields to be filled in by a user.
FIG. 5 is a block diagram of an embodiment of the microphone of the invention having an
integrated scanner. The basic design is similar to that shown in figure 3, except that there is
additionally a scanning device attached to one of the ports 505 of the USB hub. A host PC 301
having a USB root hub 302 may be connected to the microphone 303 through a USB connector
305. The USB connector connects to a USB hub controller 304 in the microphone 503. The
USB hub controller 304 in the microphone 503 can interpret the various signals coming from the
elements on board the microphone 303 in order to send those signals in a meaningful way to the
host PC 301. The USB hub controller 304 thus allows multiple elements in the microphone 303
to send signals to the host PC without the signals becoming intractably entangled. In a preferred
embodiment, the USB hub controller 304 is comprised of a Texas Instruments™ TUSB2036 2-
or 3-port USB hub chip.
The scan engine 520 may be an integrated device such as the SE923-I00A, which has an
RS232 output. The output from the scan engine should be converted from the RS232 standard to
the USB standard, for example, by use of an RS232 to USB converter 510 such as an
FT8U232AM chip. In a preferred embodiment, the USB streaming controller 312 is attached to
a current sensing device or other sensor 321. When the sensor 321 senses an overcurrent or the
activation of an element with excessive current requirements, for example, when the scanning
element 520 is activated, the USB streaming controller 312 can send a signal 322 to the speaker
amplifier 320 that mutes the speaker 315, thus reducing the current requirements of the handheld
device.
FIG. 6 is a block diagram of one embodiment of the dictation device of the invention with
fingerprint security and memory stick features. The basic design of the microphone unit 603 is
similar to that shown in figure 3, except that there is an additional USB hub controller 610. The
additional USB hub controller may be part of the microphone unit itself, or it may be part of a
separate base unit 615 that is attached to the computer 301. The remainder of the description of
this embodiment presumes that there is a separate base unit 615 attached to the computer through
the USB root hub 302. I n this embodiment, additional s upply v oltage regulators 650 and an
additional power switch 620 provides power to the base unit 615.
The base unit may further have an additional USB hub controller 610. One part of the
additional USB hub controller 610 may be connected to the USB hub controller for the
microphone and pointing device 304, as described above in the text accompanying figure 3. The
other two ports of the additional USB hub controller 610 may be used for other features. In this
embodiment, one of the other two ports of the additional USB hub controller 610 may
accommodate additional memory, for example, through a memory stick such as SmartMedia™
or any other memory device, so that dictation may be digitally stored within the microphone 603, base unit 615, or both. The remaining port of the additional USB hub controller 610 may accommodate a fingerprint identification device 640. A fingerprint identification device 640 would only allow authorized users to use a microphone of the invention to dictate and sign medical reports.