RU2320082C2 - Method and device for providing a text message - Google Patents

Abstract

FIELD: engineering of receiver-transmitters with possible presentation of text message with utilization of voice signal.

SUBSTANCE: in accordance to the method, trustworthiness estimate is computed, which is used to estimate the quality of correspondence between user-determined template of message and vectors of signs of received fragment of speech, received by means of digitization and segmentation of a fragment of speech; estimate of trustworthiness is compared to set threshold value and if the trustworthiness estimate exceeds the set threshold value, then the template of the message is added to the list of selected message templates, in opposite case the template is declined; and the text message is presented to user in accordance to the list of selected message templates.

EFFECT: simplified composition of short text messages and increased personification of same.

2 cl, 4 dwg

Description

FIELD OF THE INVENTION

The invention relates to a method and apparatus for providing a text message using a voice signal. The invention, in particular, is useful, but not necessary as a limitation, for providing a text message using voice input processed by a portable electronic device having limited memory and computing capabilities.

BACKGROUND

Short text messaging, often using the Short Message Service (SMS) format, is a very popular wireless application. Billions of short text messages are sent each month, usually from one mobile phone to another. Such text messages are popular for several reasons. Messages, as a rule, make up a fraction of the cost of a one-minute call to a mobile phone, and they do not require a dial tone for sending or receiving. Therefore, messages can be created and sent at a point in time that is convenient for the sender, and received and read at a point in time that is convenient for the recipient.

Text messages are usually created by typing characters on the keypad of a mobile phone. However, the use of such small-sized keypads with a non-standard key layout for composing a message can be inconvenient and, as a rule, requires more time than might be required when using a full-sized keyboard with a standard key layout. But it is undoubtedly impractical to have a full-sized keyboard attached to a mobile phone. Thus, there is a need for a more efficient way to compose short text messages.

In addition, although various types of speech recognition systems are well known, most are not suitable for use in portable electronic devices such as mobile phones. This is because prior art speech recognition systems typically require more processing power and memory than are available in portable electronic devices.

Disclosed in the prior art vocabulary systems and methods for speech recognition use a predefined, fixed vocabulary list. In use, a fixed vocabulary list can be large, but cannot be exhaustive, and therefore, for example, the name of the person and the names of many locations may not be included. In contrast, open vocabulary systems and speech recognition methods comprise a variable vocabulary to which new words and phrases can be added by the user or by someone else. However, modern open vocabulary systems and speech recognition methods require relatively large computational overheads, which may be unacceptable for portable electronic devices such as personal digital assistants, cordless telephones and other portable devices.

In this patent description, including in the claims, the terms “comprises”, “comprising” or similar terms are intended to mean non-exclusive inclusion, so that a method or device that contains a list of elements does not exclusively include these elements, but rather may in fact include other items not listed.

SUMMARY OF THE INVENTION

In accordance with one aspect of the invention, a method for providing a text message is provided. The method includes the steps of receiving a piece of speech at the input of an electronic device. Then, a speech recognition is performed on the speech fragment, controlled by user-defined message templates stored in the memory associated with the electronic device, and the speech recognition is determined by matching the speech fragment with one of the templates to create a mapping template. Then, a text message is provided by the pattern matching.

At least one of the message templates may include a fixed language component.

At least one of the message templates may include a mutable language component.

At least one of the message templates may include both fixed and variable language components.

The text message may be an SMS message.

The above method may also include the step of editing a user-defined message template by receiving dialed characters from a keypad of an electronic device.

The text message component may be a transcription of a piece of speech.

The entire totality of a text message can be a transcription of a fragment of speech.

According to another aspect of the invention, an electronic device for providing a text message is provided. The device includes a microphone operable to receive a portion of speech; non-volatile memory for saving message templates; and a processor operable to perform speech recognition of the speech fragment driven by the message patterns, wherein the processor is operable to match the speech fragment to one of the patterns to create a matching pattern to provide a text message by the matching pattern.

As for the electronic device, message templates may also include fixed or variable language components, or both, fixed and variable, language components.

As for the electronic device, the text message may be an SMS message.

The electronic device may include a keypad operable to edit a message template.

An electronic device may be operable to match a speech fragment to a large number of patterns and calculate a likelihood score for each of the patterns.

BRIEF DESCRIPTION OF THE DRAWINGS

So that the invention can be easily understood and put into practice, reference will now be made to preferred embodiments, which are illustrated with reference to the accompanying drawings, in which:

Figure 1 is a schematic structural diagram of a radiotelephone in accordance with the present invention;

2 is a flowchart illustrating a method for providing, editing, and transmitting a text message in accordance with the present invention;

Figure 3 is a flowchart that illustrates a method for providing a list of selected message templates to a user in accordance with the present invention; and

4 is a flowchart illustrating a method for enabling a user to edit existing message templates and save new templates in static program memory in accordance with the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

With reference to FIG. 1, a radiotelephone 100 is illustrated comprising a radio frequency communication unit 105 connected to be in communication with a processor 110. The input / output (I / O) interfaces in the form of a display 115, a keypad 120, a microphone 190, and a speaker 195 also connected to be in communication with the processor 110.

The processor 110 comprises an encoder / decoder 125 with an associative read only memory (ROM) storing data for encoding and decoding voice or other signals that may be transmitted or received by the radiotelephone 100. The processor 110 also includes a microprocessor 135 connected via a common data bus 140 and an address to an encoder / decoder 125 and an associated symbolic read-only memory (ROM) 145, random access memory (RAM) 150, a static program uemoy memory 155 and a removable SIM-module 160 (subscriber identity module). Static programmable memory 155 and SIM module 160, each can store, among other things, selected incoming text messages, a telephone directory database and, as described in more detail below, outgoing text message templates. The microprocessor 135 has ports for connection to a keypad 120, a display 115, and an alarm module 165, which typically includes a speaker, a vibrator motor, and associated drives. The character read-only memory 145 stores a code for decoding or encoding text messages that can be received by the communication unit 105 input from the keypad 120.

The radio frequency communication unit 105 is a combined receiver and transmitter having a common antenna 170. The communication unit 105 comprises a transceiver 175 connected to the antenna 170 through an RF amplifier 180. The transceiver 175 is also connected to a combined modulator / demodulator 185, which connects the communication unit 105 to the processor 110 .

With reference to FIG. 2, a flowchart is presented illustrating one embodiment of the present invention, including a method 200 for providing, editing, and transmitting a text message using a radiotelephone 100. The method 200 is activated in an initial step 205. In step 210, a speech fragment is received in an input device, such as a microphone 190, of the telephone 100. Then, the processor 110 performs sampling and digitization of the vibrational signal of the speech fragment at step 215, then segmentation in step 220, before processing, to provide feature vectors representing the waveform in step 225. It should be noted that steps 215, 220, and 225 are well known in the art and therefore do not need to be explained in detail.

Then, at step 230, speech recognition is performed by feature vectors resulting from step 225. Speech recognition is controlled by user-defined message templates stored in the static program memory 155 of device 100. Message templates are described in more detail later in this patent description. The method 200 then provides a text message to the user at step 235. The message can be provided to the user using one of the I / O interfaces, such as display 115 or speaker 195, of device 100.

After the message is provided to the user, the user is then able to decide whether to edit the message in step 240. If the user decides not to edit the message, the message is transmitted in step 245 in a message format such as SMS. However, if at step 240 the user decides to edit the message, the message is edited at step 250 before being sent at step 245. In various embodiments of the present invention, the user can edit the message in several different ways, including making editorial changes to microphone 190 or a set of editorial changes on the keypad 120. Then, the method 200 ends at step 255.

In an alternative embodiment of the present invention, after the speech recognition step 230 described above, the text message providing step 235 may include providing the telephone user 100 with a list of selected message templates from which the user can select the template that is most suitable for the implied text message.

3 is a flowchart that illustrates a method 300 for providing such a list of selected templates to a user. The method 300 is activated at the initial step 305 when the user enters a command from the keypad 120 or into the microphone 190. The method 300 first uses a processor 110 that selects in step 310 a message template from the list of available message templates. Then, in step 315, the selected pattern is compared with the feature vectors provided in step 225 of method 200. The processor 110 then calculates a likelihood score in step 320, which estimates the quality of correspondence between aspects of the selected pattern and feature vectors of the input speech fragment.

Then, the processor 110 determines in step 325 whether the likelihood estimate is higher than the set threshold. The threshold value can be automatically calculated by the processor 110, or it can be pre-set by the user of the telephone 100. If the likelihood estimate of the selected template is lower than the set threshold, the template is rejected at step 330. However, if the likelihood estimate of the selected template is higher than the set threshold, then at 335 the template is considered an acceptable match to the entered speech fragment, and the template is added to the list of selected templates. Regardless of whether the selected template is rejected or added to the list of selected templates, method 300 then proceeds to step 340, where processor 110 determines whether all available templates have been evaluated. If not all available templates have been evaluated, at step 345, method 300 selects the next message template and returns to step 315, where the next template is compared with the feature vectors of the inputted speech fragment. If all the templates were evaluated in step 340, the method 300 continues to step 350 and provides the user with a list of all selected templates. Selected templates may be provided to the user using one of the I / O interfaces, such as display 115 or speaker 195, of device 100. Then, method 300 ends at block 355.

According to one embodiment of the present invention, users of the telephone 100 are not limited to using templates provided by the manufacturer of the telephone 100. Rather, users of the telephone 100 are able to edit existing templates stored in the static program memory 155 to create their own personalized message templates.

With reference to FIG. 4, a method 400 is illustrated for enabling a user to edit existing patterns and save new patterns in a static program memory 155. The method 400 is activated at an initial step 405 when a user inputs a command from a keypad 120 or into a microphone 190. At step 410 a list of existing templates is provided to the user of the phone 100 via I / O interfaces, such as a display 115 or speaker 195. The user then selects the desired message template in step 415 using the I / O interface a face, such as a microphone 190 or a keypad 120. The user then edits the template in step 420, again using an I / O interface such as a microphone 190 or a keypad 120. Finally, in step 425, the user saves the edited template in a static programmable memory 155. Then, method 400 ends at step 430.

Other methods for editing message templates are also within the scope of the present invention, including connecting the telephone 100 to a host computer using a communication channel such as a USB cable, and then downloading or group writing the edited templates to a static programmable memory 155.

The method of the present invention may further include message templates that contain fixed and mutable language components. Fixed language components do not change when the user selects a template and transmits a message. However, mutable language components can be changed by the user from message to message. The use of fixed and variable language components can greatly enhance the limited processing power and memory of the phone 100.

For example, a specific short text message template regarding a business meeting request might contain the following: “ Meet me at $ PLACE AT $ TIME.” Here, fixed language components are underlined, and mutable language components are capitalized and begin with "$". Different users of the template may then edit a variable, such as $ PLACE, to bring it into line with their specific circumstances. For example, a university student may define the $ PLACE variable as:

$ PLACE = sp | library | dormitory | cafeteria, etc.

Whereas a lawyer can define the variable $ PLACE as:

$ LOCATION = sp | office | courthouse | house, etc.

In the above, “sp” means the pause or absence of a voice event, and “|” means the logical operator “OR”.

Another example of a message template that can be used in the present invention is “ Happy $ HOLIDAY”. Here the mutable language component $ HOLIDAY can be edited by the user to include:

$ FESTIVAL = sp | birthday | new year | thanksgiving, etc.

Using open vocabulary speech recognition, the telephone 100 is capable of recognizing edited, mutable language components entered by a user. Since mutable language components consist of discrete sets of variables, the overhead of speech recognition processing and memory requirements are minimized. Thus, the foregoing is particularly suitable for devices having limited computing and storage resources, such as mobile phones.

The use of templates that include fixed and variable language components increases the effectiveness of the speech recognition system for several reasons. First, the fixed language components of a particular pattern can usually be recognized quickly and efficiently, since there are only a limited number of patterns stored in static programmable memory 155, compared to the almost unlimited number of phrase permutations associated with natural language phrase structures . Secondly, mutable language components can also be effectively recognized, since the position within the phrase of the mutable language component in the message template automatically identifies a discrete set of possible answers. For example, with reference to the Happy $ FESTIVAL message template above, the fixed Happy language component can act as a signal so that processor 110 is aware that the next voice input received from microphone 190 will be a mutable language component $ HOLIDAY. "

Although the foregoing embodiments of the present invention have been described with respect to the radiotelephone 100, the method and apparatus of the present invention could also include other electronic devices that provide text messages, such as personal digital assistants (PDAs).

Thus, the present invention simplifies the steps required to provide and transmit a text message from a portable electronic device. A text message can be provided via voice input rather than typed characters entered from a small keypad. In addition, the invention may include open vocabulary speech recognition to avoid the increased memory requirements of prior art vocabulary speech recognition. Open vocabulary speech recognition uses speaker-independent acoustic models of parts of a word designed to cover all acoustic cases of use, or phonemes, of a language. Thus, the user is not limited to a predefined dictionary, but can edit mutable language components as described above to include words not found in the dictionary, such as names and locations. The result is that the text messages provided by the present invention can be highly personalized.

The above detailed description provides only preferred exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the invention. A more detailed description of the preferred exemplary embodiments provides those skilled in the art with sufficient description to implement the preferred exemplary embodiments of the invention. It should be clear that various changes can be made in the operation and layout of the elements without leaving the essence and scope of the invention as set forth in the attached claims.

Claims (13)

1. The method of providing a text message in which receive a piece of speech at the input of an electronic device; the said speech fragment is compared with one of the user-defined message templates stored in the memory associated with the electronic device, and a likelihood score is calculated that evaluates the quality of correspondence between the said message template and the speech fragment attribute vectors obtained by sampling and segmenting the speech fragment; is the likelihood estimate higher than the set threshold, and if the likelihood estimate is higher than the set threshold about the value, then the message template is added to the list of selected message templates, otherwise the message template is rejected, and a text message is provided to the user in accordance with the list of selected message templates. 2. The method according to claim 1, in which at least one of the message templates contains a fixed language component. 3. The method according to claim 1, in which at least one of the message templates contains a mutable language component. 4. The method according to claim 1, in which at least one of the message templates contains both fixed and variable language components. 5. The method according to claim 1, wherein the text message is an SMS message. 6. The method according to claim 1, which also includes editing the said user-defined message template by receiving the dialed characters from the keypad of the electronic device. 7. The method according to claim 1, in which the component of the text message is a transcription of a fragment of speech. 8. The method according to claim 1, in which the entire contents of the text message is a transcription of a fragment of speech. 9. An electronic device for providing a text message containing a microphone for receiving a portion of speech; non-volatile memory for storing message templates; and a processor for matching said speech fragment received by the microphone with one of said message patterns stored in non-volatile memory associated with an electronic device, wherein the processor is configured to discretization and segmentation of a speech fragment to obtain vectors of signs of a speech fragment; calculating a likelihood score that evaluates the quality of correspondence between the message template and the feature vectors of the speech fragment determining whether the likelihood estimate is higher than the set threshold, and if the likelihood estimate is higher than the set threshold, then add the message template to the list of selected message templates, otherwise reject the message template; and providing the user with a text message in accordance with the list of selected message templates. 10. The device according to claim 9, in which at least one of the message templates contains a fixed language component. 11. The device according to claim 9, in which at least one of the message templates contains a mutable language component. 12. The device according to claim 9, in which at least one of the message templates contains both fixed and variable language components. 13. The device according to claim 9, in which the text message is an SMS message.

Images