Text-to-speech driven annunciation of caller identification

FIELD OF THE INVENTION

This invention relates to annunciation devices for audibly annunciating the identity of a calling party at telephone equipment connected to a telephone switch.

BACKGROUND OF THE INVENTION

Most telephone operating companies have been offering calling line identification services in which the information of a calling party's telephone number and/or name is sent from a local central office to the called party's telephone station apparatus if a called party is a subscriber of that service. Early proposed versions of this service may be found in United States Patent Nos. 3,727,003 and 3,812,296 to Paraskevakos, as well as in United States Patent No. 3,787,626 to Subieta. A more recent scheme for providing this service is discussed in United States Patent Re. 31,789 to Hashimoto.

In these patents, the directory number of a calling party is coded as FSK modulated pulse trains and is transmitted from the originating central office to the terminating office and sent to the called party's telephone station apparatus either before an alerting signal such as ringing is sent or during a silent interval between ringing signals. In these patents, the called party's station apparatus must be equipped with receiving circuitry for decoding the FSK modulated digital signal as well as a display device to indicate the identity of the calling party. The called party is then placed in the position of being given the opportunity to answer the call or ignore it. United States Patent No. 5,263,084 to Chaput et al., discloses a method and a telephone station apparatus of displaying an identification of a third calling party.

Such visible identification of the calling party's telephone number or name is not advantageous for vision impaired people nor is visible identification convenient for identifying the calling party in the case where the called party is away from the telephone station apparatus which displays a calling party's telephone number in an onhook condition. Various schemes, therefore, have been proposed for providing audible annunciation of the calling party. One such scheme is disclosed in United States Patent No. 4,922,490 to Blakely, however, this scheme involves interpretation of the calling party's identification at a central office switch and transmission of audio information directly to a telephone, the telephone being equipped with audio annunciation apparatus for producing an audio annunciation based on the audio signal provided by the central office. This renders the quality of the audio heard by the user dependent upon the audio production equipment at the central office. In addition, Magnovox has a website on the Internet at http://www.esn.net/callerid/TCmore.html which describes a voice announced caller identification unit. However, there is no indication that this device cooperates with text to speech converter or an external database to produce good quality audio. The present invention addresses this need.

The use of CLID information directly as it is provided by the telephone switch does not necessarily convert into a meaningful result when applied to a text to speech converter. For example, if the subscriber lists his name as T.A. Kuechler, the initials T.A.K may be treated as the word or syllable "tak". This may defeat the purpose of audibly announcing the caller's name as the name cannot be readily recognized as that of Mr. Kuechler. What would be desirable, therefore, is a device which would apply some intelligence to the information received as Caller Identification Information to provide an optimum audible announcement of the identity of the calling party.

SUMMARY OF THE INVENTION

In accordance with one aspect of the invention, there is provided a method of producing an audio tag in response to calling line identification information, The method includes the steps of, receiving calling line identification information from a telephone line, searching a directory, using at least some of the calling line identification information as a search parameter to locate a directory record corresponding to at least some of the calling line identification information, communicating information held in the directory record or in the calling line identification information to a text to speech converter to produce the audio tag.

Preferably, the method includes the step of playing the audio tag to identify the calling party.

Preferably, the method includes the step of searching a predefined computer readable telephone directory of names and numbers.

Preferably, the method includes the step of using a telephone number included in the calling line identification information as a search parameter to locate a name corresponding to the telephone number.

Preferably, the method includes the step of evaluating the calling line identification information and the information held in the directory record to determine which would sound best when applied to the text to speech converter.

Preferably, the method includes the step of determining the total number of characters in the calling line identification information and the information in the directory record.

Preferably, the method includes the step of determining the total number of vowels in the calling line identification information and the information in the directory record.

Preferably, the method includes the step of determining the number of vowels enclosed by consonants in the calling line identification information and the information in the directory record.

Preferably, the method includes the step of determining the total number of vowels in the calling line identification information and the information in the directory record.

Preferably, the method includes the step of determining the number of vowels enclosed by consonants in the calling line identification information and the information in the directory record.

Preferably, the method includes the step of determining which of the calling line identification information or the information in the directory record will sound best when applied to the text to speech converter by calculating respective audibility score values calculated as the weighted sum of the total number of characters, the total number of vowels and the number of vowels enclosed by consonants.

Preferably, the method includes the step of communicating to the text to speech converter, the calling line identification information or the information in the directory record according to which has the greatest number of characters, the greatest number of vowels and the greatest number of vowels enclosed by consonants.

In accordance with another aspect of the invention, there is provided a computer-readable storage medium on which is stored a plurality of computer-readable instructions for directing a processor to produce an audio tag in response to calling line identification information by receiving calling line identification information from a telephone line, searching a directory, using at least some of the calling line identification information as a search parameter to locate a directory record corresponding to at least some of the calling line identification information, communicating information held in the directory record or the calling line identification information to a text to speech converter to produce the audio tag.

In accordance with another aspect of the invention, there is provided an apparatus for producing an audio tag in response to calling line identification information. The apparatus includes a receiver for receiving calling line identification information from a telephone line, a processor for accessing and searching a directory, using at least some of the calling line identification information as a search parameter to locate a directory record corresponding to at least some of the calling line identification information, a text to speech converter for converting information held in the directory record or in the calling line identification information to produce the audio tag.

Preferably, the apparatus includes a digital to analog converter for converting the audio tag into an analog signal.

Preferably, the apparatus includes a speaker for converting the audio tag into an audible sound identifying the calling party.

Preferably, the digital to analog converter and speaker are controlled by the processor.

Preferably, the processor is programmed to search a predefined computer readable telephone directory of names and numbers.

Preferably, the processor is programmed to use a telephone number included in the calling line identification information as a search parameter to locate a name corresponding to the telephone number.

Preferably, the processor is programmed to evaluate the calling line identification information and the information held in the directory record to determine which would sound best when applied to the text to speech converter.

Preferably, the processor is programmed to determine the total number of characters in the calling line identification information and the information in the directory record.

Preferably, the processor is programmed to determine the total number of vowels in the calling line identification information and the information in the directory record.

Preferably, the processor is programmed to determine the number of vowels enclosed by consonants in the calling line identification information and the information in the directory record.

Preferably, the processor is programmed to determine which of the calling line identification information and the information in the directory record will sound best when applied to the text to speech converter by calculating respective audibility score values calculated as the weighted sum of the total number of characters, the total number of vowels and the number of vowels enclosed by consonants.

Preferably, the processor is programmed to communicate to the text to speech converter, the calling line identification information or the information in the directory record according to which has the greatest number of characters, the greatest number of vowels and the greatest number of vowels enclosed by consonants.

Preferably, the text to speech converter includes instructions for directing the processor to produce an audio tag in response to text input.

Preferably, the text to speech converter is controlled by the processor.

Preferably, the directory includes instructions for directing the processor to access a memory medium containing the directory records.

Preferably, the memory medium is located remotely from the processor.

Preferably, the memory medium is controlled by the processor.

Preferably, the directory is controlled by the processor.

In accordance with another aspect of the invention, there is provided a method of audibly identifying a calling party. The method includes the steps of receiving calling line identification information from a telephone line, searching a directory, using at least some of the calling line identification information as a search parameter to locate a directory record corresponding to at least some of the calling line identification information, communicating information held in the directory record or in the calling line identification information to a text to speech converter to produce an audio tag and playing the audio tag to identify the calling party.

In accordance with another aspect of the invention, there is provided a computer-readable storage on which is stored a plurality of computer-readable instructions for directing a processor to audibly identify a calling party by executing the method steps listed immediately above.

In accordance with another aspect of the invention, there is provided an apparatus for audibly identifying a calling party, the apparatus comprising: <ul id="ul0001" list-style="none"><li>a) receiving means for receiving calling line identification information from a telephone line;</li><li>b) searching means for searching a directory, using at least some of said calling line identification information as a search parameter to locate a directory record corresponding to said at least some of said calling line identification information;</li><li>c) text to speech conversion means for converting information held in said directory record or in said calling line identification information to produce said audio tag; and</li><li>d) audio tag conversion means for converting said audio tag into an audible sound to identify said calling party.</li></ul>

In accordance with another aspect of the invention therre is provided an apparatus for producing an audio tag in response to calling line identification information, the apparatus comprising: <ul id="ul0002" list-style="none"><li>a) receiving means for receiving calling line identification information from a telephone line;</li><li>b) searching means for searching a directory, using at least some of said calling line identification information as a search parameter to locate a directory record corresponding to said at least some of said calling line identification information; and</li><li>c) text to speech conversion means for converting information held in said directory record or in said calling line identification information to produce said audio tag.</li></ul>

Preferably the apparatus further includes means for playing said audio tag to identify said calling party.

Preferably the searching means includes means for searching a predefined computer readable telephone directory of names and numbers.

Preferably the means for searching a directory includes means for using a telephone number included in said calling line identification information as a search parameter to locate a name corresponding to said telephone number.

Preferably the apparatus further includes means for evaluating said calling line identification information and said information held in said directory record to determine which would sound best when applied to said text to speech converter.

Preferably the means for evaluating includes means for determining the total number of characters in said calling line identification information and said information held in said directory record.

Preferably the means for evaluating includes means for determining the total number of vowels in said calling line identification information and said information held in said directory record.

Preferably the means for evaluating includes means for determining the number of vowels enclosed by consonants in said calling line identification information and said information held in said directory record.

Preferably the means for evaluating includes means for determining the total number of vowels in said calling line identification information and said information held in said directory record.

Preferably the means for evaluating includes means for determining the number of vowels enclosed by consonants in said calling line identification information and said information held in said directory record.

Preferably the apparatus includes means for determining whether said calling line identification information or said information held in said directory record will sound best when applied to said text to speech converter by calculating respective audibility score values calculated as the weighted sum of said total number of characters, said total number of vowels and said number of vowels enclosed by consonants.

preferably the apparatus further includes means for communicating to said text to speech converter, said calling line identification information or said information in said directory record according to which has the greatest number of characters, the greatest number of vowels and the greatest number of vowels enclosed by consonants.

In accordance with another aspect of the present invention there is provided a method of annunciating the identity of a caller in response to calling line identification information, the method comprising the steps of: <ul id="ul0003" list-style="none"><li>a) receiving calling line identification information from a telephone line;</li><li>b) searching a personal directory of personal directory records having name, number and audio tag fields to locate a record corresponding to said calling line identification information and on finding such record annunciating the contents of said audio tag field,</li><li>c) on not finding such a record corresponding to said calling line identification information searching a second directory, having records with name and number fields, using at least some of said calling line identification information as a search parameter to locate a directory record corresponding to said at least some of said calling line identification information;</li><li>c) communicating information held in said directory record or in said calling line identification information to a text to speech converter to produce an audio tag; and</li><li>d) annunciating said audio tag.</li></ul>

BRIEF DESCRIPTION OF THE DRAWINGS

In drawings which illustrate embodiments of the invention, <dl id="dl0001"><dt>Figure <b>1</b></dt><dd>is a perspective view of an apparatus according to a first embodiment of the invention;</dd><dt>Figure <b>2</b></dt><dd>is a schematic representation of programs running on a processor of the apparatus shown in Figure <b>1</b>;</dd><dt>Figure <b>3</b></dt><dd>is a tabular representation of a directory table in a database accessed according to the first embodiment of the invention;</dd><dt>Figure <b>4</b></dt><dd>is a tabular representation of a call information structure shown in Figure <b>2</b>;</dd><dt>Figure <b>5</b></dt><dd>is a flowchart of a database access routine according to the first embodiment of the invention;</dd><dt>Figure <b>6</b></dt><dd>is a tabular representation of a database query used by a text evaluation application shown in Figure <b>2</b>;</dd><dt>Figure <b>7</b></dt><dd>is a schematic representation of a query response produced by the database accessed according to the first embodiment of the invention;</dd><dt>Figure <b>8</b></dt><dd>is a flowchart of a caller identification name evaluation routine according to the first embodiment of the invention;</dd><dt>Figure <b>9</b></dt><dd>is a schematic representation of registers used in evaluating information received from the subscriber line and information contained in a database record; and</dd><dt>Figure <b>10</b></dt><dd>is a flowchart of a best name routine according to the first embodiment of the invention.</dd></dl>

DETAILED DESCRIPTION

Referring to Figure <b>1</b>, an apparatus for producing an audio tag in response to calling line identification information and for audibly identifying a calling party according to a first embodiment of the invention is shown generally at <b>10</b>. In this embodiment, the apparatus includes a personal computer <b>12</b> having a processor <b>14</b> in communication with a display <b>16</b>, a keyboard <b>18</b>, a floppy disk drive <b>20</b>, for receiving a floppy disk <b>25</b>, a CD ROM drive <b>22</b> and a hard disk drive <b>24</b>. In addition, further connected to the processor <b>14</b> is a modem card <b>26</b> having a subscriber line interface <b>28</b> for connection to a subscriber line <b>30</b> connected to a telephone company switch (not shown). The modem card <b>26</b> and subscriber line interface act as a receiver and receiving means for receiving calling line identification information from a telephone line.

The personal computer <b>12</b> also has a sound card <b>29</b> having a digital to analog converter <b>31</b> operable to provide signals to a speaker <b>33</b> on the personal computer <b>12</b>.

Figure <b>2</b>

Referring to Figure <b>2</b>, the personal computer <b>12</b> is controlled by a multitasking operating system <b>32</b>, which in this embodiment, is Windows<b>95</b> (tm) produced by Microsoft, Inc. The multitasking operating system has a telephony service provider module <b>34</b> which communicates with the modem card <b>26</b> shown in Figure <b>1</b> to receive telephone caller identification information (CLID) from the subscriber line <b>30</b>. Referring back to Figure <b>2</b>, the multitasking operating system further includes a Telephone Application Programmer's Interface (TAPI) <b>36</b> and a call information structure table <b>38</b>, the call information structure table <b>38</b> being produced by the telephone application programmer's interface <b>36</b>.

The multitasking operating system is resident on the hard disk drive <b>24</b> shown in Figure <b>1</b>. In addition, a database program <b>40</b> such as SelectPhone (tm) is installed on the hard disk drive <b>24</b> and a database of callers associated with the database program <b>40</b> is accessible through the CD ROM drive <b>22</b> which provides access to a memory medium, in particular a CD ROM provided by the database manufacturer. Thus, the memory medium is controlled by the processor.

Figure <b>3</b>

Referring to Figure <b>3</b>, the CD ROM contains a plurality of directory tables <b>41</b> having directory records <b>43</b> with directory number fields <b>45</b> and directory first name, last name fields <b>47</b>. Each record has a ten digit telephone number of a subscriber stored in the directory number field <b>45</b> and has a subscriber name stored in the directory first name, last name field <b>47</b>. In this embodiment, the directory table includes records of all telephone subscribers in the United States. Thus, the directory is controlled by the processor.

Referring back to Figure <b>2</b>, the personal computer <b>12</b> further includes a text evaluation application program <b>42</b> and a text to speech converter program <b>44</b>. The text to speech converter program is of the type provided by Berkeley Speech Systems of Berkeley, California and essentially receives textual input and produces signals for controlling the sound card <b>29</b> to produce audible signals through the digital to analog converter <b>31</b>, which are applied to the speaker <b>33</b>.

Figure <b>4</b>

Referring to Figure <b>4</b>, the call information structure table is shown generally at <b>38</b> and includes a long-distance flag field <b>46</b>, a CLID name field <b>48</b>, a CLID number field <b>50</b>, a time of day field <b>52</b>, a name privacy flag field <b>54</b>, a number privacy flag field <b>56</b> and an unknown caller flag field <b>58</b>. The above fields are for storing variables by the same names received from the calling line identification information received by the telephony service provider interface <b>34</b> from the modem card <b>26</b> and subscriber line <b>30</b>. The TAPI <b>36</b> formats the calling line identification information received from the subscriber line into the format shown in Figure <b>4</b>.

Referring to Figures <b>1</b> and <b>2</b>, the text evaluation application <b>42</b> includes computer-readable instruction codes stored on a floppy disk, which are transferred by the processor from the floppy disk <b>25</b> to the hard disk drive <b>24</b>. These codes are operable to direct the processor <b>14</b> to execute a database access routine, a caller identification name evaluation routine and a best name routine as shown in Figures <b>5</b>, <b>8</b> and <b>10</b> respectively.

Referring to Figure <b>5</b>, the text evaluation application begins with the database access routine <b>68</b> which begins with a first block <b>70</b> which directs the processor to execute a database open command on the SQL interface of the SelectPhone application. Referring to Figures <b>5</b> and <b>6</b>, block <b>72</b> then directs the processor to compose a structured query in the form shown. The CLID number is copied into the query from the CLID number buffer <b>50</b>. The processor then executes block <b>74</b> which directs it to access the database program <b>40</b> shown in Figure <b>2</b> to make the query produced in block <b>72</b>. Thus, the processor is programmed to act as means for searching a predefined computer readable telephone directory of names and numbers.

Referring to Figures <b>3</b> and <b>7</b>, on accessing the database, if a directory record <b>43</b> is found in the directory table <b>41</b>, the record having directory number field <b>45</b> contents equal to the telephone number identified in the CLID number field <b>50</b> of the call information structure table shown in Figure <b>4</b>, the contents of the first name, last name field <b>47</b> are copied to a query response buffer <b>66</b> shown in Figure <b>7</b>. Thus, the processor acts as means for accessing and searching a directory, using at least some of the calling line identification information as a search parameter to locate a directory record corresponding to at least some of the calling line identification information. More particularly, the processor is programmed to act as means for using a telephone number included in the calling line identification information as a search parameter to locate a name corresponding to the telephone number.

Referring back to Figure <b>5</b>, block <b>76</b> then directs the processor <b>14</b> to determine whether or not the query response buffer <b>66</b> is empty or, in other words, whether or not it contains non-zero contents. If the query response buffer <b>66</b> is empty, ie., has zero contents, the processor is directed to the caller identification name evaluation routine shown in Figure <b>8</b>. Otherwise, block <b>78</b> directs the processor to calculate the total number of characters of the contents of the query response buffer <b>66</b> shown in Figure <b>7</b> and to store the value so calculated in a D₁ register <b>80</b> shown in Figure <b>9</b>. Thus, the processor is programmed to act as determining means for determining the total number of vowels in the calling line identification information and the information held in the directory record.

Referring back to Figure <b>5</b>, block <b>82</b> then directs the processor to calculate the total number of vowels in the name field <b>66</b> and store the number so calculated in a D₂ register <b>84</b> shown in Figure <b>9</b>. Thus, the processor is programmed to act as means for determining the total number of vowels in the calling line identification information and the information held in the directory record.

Referring back to Figure <b>5</b>, block <b>86</b> directs the processor to calculate the greatest number of vowels enclosed by consonants in the query response buffer <b>66</b> and to store the number so calculated in a D₃ register <b>88</b> shown in Figure <b>9</b>. Thus, the processor is programmed to act as means for determining the number of vowels enclosed by consonants in the calling line identification information and the information held in the directory record.

Referring back to Figure <b>5</b>, block <b>90</b> directs the processor to calculate an audibility score as the weighted sum of contents of the D₁, D₂ and D₃ registers <b>80</b>, <b>84</b> and <b>88</b> shown in Figure <b>9</b>. In this regard, respective weights W₁ 92, W₂<b>94</b> and W₃<b>96</b> are applied to the contents of registers D₁<b>80</b>, D₂<b>84</b> and D₃<b>88</b> to produce the audibility score S1 according to the relation S<b>1</b> = W₁D₁ + W₂D₂ + W₃D₃, and the resulting score S<b>1</b> is stored in register S<b>1 98</b> shown in Figure <b>9</b>.

Referring back to Figure <b>5</b>, the database access routine is thus completed and the processor is directed to the caller identification name evaluation routine shown in Figure <b>8</b>.

Referring to Figure <b>8</b>, the caller identification name evaluation routine is shown generally at <b>100</b> and begins with a first block <b>102</b> which directs the processor to determine whether or not the contents of the CLID name field <b>48</b> shown in Figure <b>4</b> are blank. If so, the processor is directed to the best name routine shown in Figure <b>10</b>. Otherwise, block <b>104</b> directs the processor to calculate the total number of characters in the CLID name stored in the CLID name field <b>48</b> of Figure <b>4</b> and store the result in a C₁ register <b>106</b> shown in Figure <b>9</b>. Thus, the processor is programmed to act as means for determining the total number of vowels in the calling line identification information and the information held in the directory record.

Referring back to Figure <b>8</b>, block <b>108</b> then directs the processor to calculate the total number of vowels in the contents of the CLID name field <b>48</b> shown in Figure <b>4</b> and to store the number so calculated in a C₂ register <b>110</b> shown in Figure <b>9</b>.

Referring back to Figure <b>8</b>, block <b>111</b> then directs the processor to calculate the greatest number of vowels enclosed by consonants and to store the result so calculated in a C₃ register <b>112</b> shown in Figure <b>9</b>. Thus, the processor is programmed to act as means for determining the number of vowels enclosed by consonants in the calling line identification information and the information held in the directory record.

Referring back to Figure <b>8</b>, block <b>114</b> then directs the processor to calculate an audibility score W₂ for the CLID information, as a weighted sum of the contents of registers C₁<b>106</b>, C₂<b>110</b> and C₃<b>112</b>, according to the relation S<b>2</b> = W₁C₁ + W₂C₂ + W₃C₃, where the contents of weighting registers W₁<b>92</b>, W₂<b>94</b> and W₃<b>96</b> shown in Figure <b>9</b> are used as weighting factors. The audibility score W₂ is stored in the audibility score register S<b>2 116</b>.

Thus, the processor is programmed to act as means for determining which of the calling line identification information and the information held in the directory record will sound best when applied to the text to speech converter by calculating respective audibility score values calculated as the weighted sum of the total number of characters, the total number of vowels and the number of vowels enclosed by consonants.

Referring back to Figure <b>6</b>, the processor is then directed to the best name routine shown in Figure <b>10</b>.

Figure <b>10</b>

Referring to Figures <b>9</b> and <b>10</b>, the best name routine is shown generally at <b>118</b> and begins with a first block <b>120</b> which directs the processor to determine whether or not the contents of the S<b>1</b> and W₂ registers <b>98</b> and <b>116</b> are equal to zero. If so, the best name routine is ended and no sound or audible indication of the caller's name is produced by the device.

If the contents of the S<b>1</b> and W₂ registers <b>98</b> and <b>116</b> are not equal to zero, block <b>122</b> directs the processor to determine whether or not the contents of the S<b>1</b> register <b>98</b> are greater than or equal to the contents of the W₂ register <b>116</b>. If so, then block <b>124</b> directs the processor to communicate the contents of the query response buffer <b>66</b> to the text to speech converter program <b>44</b> in Figure <b>2</b>. The text to speech converter program then produces an audio tag in response to the contents of the query response buffer <b>66</b> and such audio tag is applied to the digital to analog converter <b>31</b> on the sound card <b>29</b> to cause the speaker <b>33</b> to audibly annunciate the contents of the query response buffer <b>66</b> shown in Figure <b>3</b>. Thus, the text to speech converter acts as a means for converting information held in the directory record or in the calling line identification information to produce an audio tag, the a digital to analog converter acts as a means for converting the audio tag into an analog signal and the speaker acts as a means for converting the audio tag into an audible sound identifying the calling party. It will be appreciated that the digital to analog converter and speaker are controlled by the processor. In effect, the digital to analog converter and speaker act as an audio tag converter or audio tag conversion means for playing the audio tag to identify the calling party.

Referring back to Figure <b>10</b>, if at block <b>122</b>, the contents of the S<b>1</b> register <b>98</b> are not greater than or equal to the contents of the W₂ register <b>116</b>, block <b>126</b> directs the processor to communicate the contents of the CLID name field <b>48</b> shown in Figure <b>4</b> to the text to speech converter program <b>44</b> shown in Figure <b>2</b>. The text to speech converter program <b>44</b> then produces a digital signal representing an audio tag representing the contents of the CLID name field <b>48</b> shown in Figure <b>4</b>. The text to speech converter includes instructions for directing the processor to produce an audio tag in response to text input and the text to speech converter is controlled by the processor. The audio tag is then presented to the digital to analog converter <b>31</b> which produces an analog signal to drive the speaker <b>33</b> to cause the contents of the CLID name field <b>48</b> to be audibly annunciated. In effect, therefore, the processor is programmed to acts as a means for evaluating the calling line identification information and the information held in the directory record to determine which would sound best when applied to the text to speech converter. More particularly, the processor is programmed to communicate to the text to speech converter, the calling line identification information or the information held in the directory record according to which has the greatest number of characters, the greatest number of vowels and the greatest number of vowels enclosed by consonants.

Alternatives

Alternatively, the database program <b>40</b> may be replaced with a communications program which automatically accesses a remotely located database on a remotely located memory medium through a further communications line (not shown), which may include a further subscriber line which may include a network to which the personal computer <b>12</b> is connected. In this manner, a plurality of personal computers <b>12</b> may share the same database for locating directory records to determine the best representation for annunciation.

Alternatively, referring to Figure <b>1</b>, the processor is located inside an intelligent telephone rather than the personal computer <b>12</b>. In such an embodiment, the text to speech converter may include a digital signal processor under the control of the processor <b>14</b> for producing audio tags in response to applied text as described above. Also, in such an embodiment, the directory which was described to have been stored on CD ROM in connection with the first embodiment, would be stored in memory within the intelligent telephone.

In a further embodiment of the invention, referring to Figure <b>1</b>, the user may maintain in the hard disk drive <b>24</b>, a personal directory comprised of names, telephone numbers and voice tags which are prerecorded by the user. In addition, there may also be stored on the hard disk drive <b>24</b>, a company phone list of extension numbers and names. The personal directory and company directory are used in addition to the directory table described above including the names and numbers of all telephone subscribers in the United States.

In this further embodiment, the database access routine is amended to include, preceding block <b>70</b>, a query of the personal directory to locate a corresponding record containing a name of a caller and routines for calculating a score as described above associated with the name stored in the personal directory. In addition, the database access routine includes instructions for directing the processor to query the company phone list for a name and on finding such name, also calculating an audibility score. The modifications further include testing the score associated with the record found in the personal directory for a value greater than a preset threshold value and if such preset threshold value is exceeded, causing the contents of the name field of the personal directory to be applied to the text to speech converter. If, however, the name field in the personal directory has a score below the predefined value, the processor is directed to query the company phone list where the score associated with the name located in that list is compared against the threshold value and if above the threshold value the name associated with the company phone list is forwarded to the text to speech converter. If that name, however, has a score below the threshold value then the processor is directed to the database access routine <b>68</b> shown in Figure <b>5</b> and processing continues as indicated with respect to the first embodiment of the invention.

It will be appreciated that in general, the apparatus receives calling line identification information and uses such information to look up in a directory a corresponding entry, which may have a name associated with the number provided by the calling line identification information. The apparatus then evaluates the calling line identification information and the name associated with the record located in the directory to determine which would sound best when presented to the text to speech converter. The expected best information is then applied to the text to speech converter to produce an audible indication of the party calling.

While specific embodiments of the invention have been described and illustrated, such embodiments should be considered illustrative of the invention only and not as limiting the invention as construed in accordance with the accompanying claims.