Method and apparatus for identifying words entered on DTMF pushbuttons

• US 5,031,206 A
• Filed: 07/28/1988
• Issued: 07/09/1991
• Est. Priority Date: 11/30/1987
• Status: Expired due to Fees

• Current Assignee: Fon-Ex Incorporated
• Original Assignee: Fon-Ex Incorporated
• Inventors: Riskin, Bernard N.
• Primary Examiner(s): Oberley, Alvin E.
• Assistant Examiner(s): HJERPE, RICHARD A
• Application Number:
  US07/225,266
• Time in Patent Office: 1,076 Days
• Field of Search: 379/88, 379/89, 379/93, 379/96, 379/97, 379/98, 379/113, 379/354, 382/40
• US Class Current: 379/93.27
• CPC Class Codes: H04M 11/066 Telephone sets adapted for data transmision
• CPC Class Codes: Y10S 379/906 Touchtone message transmission
• CPC Class Codes: Y10S 379/908 Multimedia

Abstract

The identity of a word entered on DTMF (Dual Tone Multi-Frequency) pushbuttons is determined by identifying letters according to their frequency in groups of 2s and 3s and up to n-letter groups. Initially, a sender will type a word on a DTMF pushbutton pad. Since each button represents three (3) possible letters, or four in the case of 7 (PQRS) and 9 (WXYZ), the system will look up the possible meanings for the word from an internal memory which includes a dictionary. If the word does not exist in the dictionary, then the system will search against a Forward and a Reverse Dictionary to guess at the identity of fragments from the beginning and ending of the word. Then it will guess the identity of each letter according to its relative frequency in groups of 3s known as Trigrams, or in groups of 2s known as Digrams. The Trigams are preferably arranged in five (5) files to identify the first, second, third and final letters of a word as well as any non-specified middle location. The system can be extended to any letter grouping of size n where n is two or more. Once identified, the word is stored as part of a message and then transmitted as ASCII digital information over a packet data network to a remote receiver where digital to voice synthesis converts the message into audio. Other delivery methods include alphanumeric pagers, FAX, E-Mail, telex, computer printed output, telegrams and cables.

Claims

1. An apparatus for determining the identity of an input word of a specific language entered on a DTMF keyboard as a string of numbers, said apparatus comprising:

a full word dictionary means for storing common words and the DTMF numeric strings corresponding to said common words;

a forward dictionary means for storing full words and parts of words and the DTMF numeric strings corresponding to said full words and said parts of words, wherein said full words and said parts of words are selected and stored according to the length and frequency of occurrence in said specific language of said full words and said parts of words and wherein the letters of said full words and said parts of words are in their natural letter sequences;

a reverse dictionary means for storing reversed strings formed by reversing the natural letter sequences of said full words and said parts of words and the DTMF numeric strings corresponding to said reversed strings, wherein said reversed strings and the DTMF numeric strings corresponding to said reversed strings are stored according to the length and frequency of occurrence in said language of said reversed strings;

means for comparing the string of numbers corresponding to said input word with the numeric strings in said full word dictionary means; and

means for comparing the string of numbers corresponding to said input word with the DTMF numeric strings corresponding to said full words and said parts of words in said forward dictionary means and with the DTMF numeric strings corresponding to said reversed strings in said reverse dictionary means,

wherein said string of numbers corresponding to said input word is first compared against said DTMF numeric strings in said full word dictionary means and compared against said DTMF numeric strings corresponding to said full words and said parts of words in said forward dictionary means and said DTMF strings corresponding to said reversed strings in said reverse dictionary means if said input word is not found in said full word dictionary means.

2. The apparatus of claim 1 further comprising:

letter subgroup examining means for identifying each letter of said input word by examining said string of numbers, said letter subgroup examining means comprising means to divide said string of numbers into groups of n numbers, means for storing snippets comprising n letters and numeric strings corresponding to said snippets, said snippets being selected and stored according to the frequency of said snippets in said language, means for comparing said groups of n numbers with said numeric strings corresponding to said snippets, wherein n is an integer and equals at least 2, and

wherein said string of numbers corresponding to said input word is examined by said letter subgroup examining means if said input word is not found in said full word dictionary means, said forward dictionary means or said reverse dictionary means.

3. The apparatus of claim 2 wherein said letter subgroup examining means includes:

an n-gram-initial file for identifying the first letter of said input words;

an n-gram-2 file for identifying the second letter of said input words;

an n-gram-3 file for identifying the third letter of said input words;

an n-gram-M file for identifying any letter that is not the first, second, third or last letter of said input words; and

,

an n-gram-F file for identifying the final letter of said input words.

4. The apparatus of claim 3 further comprising:

at least one digram file for storing in groups of 2, letters according to the frequency of occurrence in said specific language of said groups of 2 letters and for storing the DTMF numbers corresponding to said groups of 2 letters.

wherein groups of two numbers of said numeric string corresponding to said input word are compared against said DTMF numbers corresponding to said groups of 2 letters in said digram file.

5. The apparatus of claim 1 further comprising:

pager means for receiving a message comprising input words identified by said apparatus.

6. The apparatus of claim 1 further comprising:

voice mail means for receiving messages comprising input words identified by said apparatus.

7. The apparatus of claim 1 further comprising:

data base means interrogatable by said apparatus for providing data based upon input words identified by said apparatus.

8. The apparatus of claim 1 further comprising:

printer means for printing input words identified by said apparatus,

whereby messages printed on said printer means are deliverable by the postal service.

9. The apparatus of claim 1 further comprising;

E-mail means for receiving messages comprising input words identified by said apparatus and for transmitting same to remote recipients.

10. The apparatus of claim 1 further comprising:

telex means for receiving messages comprised of input words identified by said apparatus and for transmitting said messages to remote recipients.

11. The apparatus of claim 1 further comprising:

cable means for receiving message comprised of input words identified by said apparatus and for transmitting said messages to remote recipients.

12. The apparatus of claim 1 further comprising:

FAX means for receiving message comprised of input words identified by said apparatus.

13. The apparatus of claim 1 further comprising:

message storing means for storing messages comprised of words identified by said apparatus,

wherein said apparatus compares number strings against said message storing means in order to identify subsequent occurrences of said number string.

14. A method for determining the identity of an input word of a specific language, said method comprising the steps of:

entering said input word on a DTMF push button pad where at least one push button represents more than one letter and said push button is identified by one and only one number;

forming a string of numbers corresponding to said input word from said letters entered on said DTMF push button pad;

comparing said string of numbers corresponding to said input word against DTMF numeric strings corresponding to words stored in a word guessing dictionary means to determine the identity of said input word; and

,

if said input word is not found in said word guessing dictionary means then comparing said string of numbers corresponding to said input word against DTMF numeric strings corresponding to words and parts of words stored in a forward dictionary means wherein said words and parts of words are stored according to the length and frequency of use of said words and parts of words in said specific language and wherein the letters of said words and parts of words are in their natural sequence, and also comparing said string of numbers corresponding to said input word against DTMF numeric strings corresponding to reversed words stored in a reverse dictionary wherein said reversed words are formed by reversing the natural sequence of the letters of said words and said parts of words and wherein said reversed words are stored according to the length and frequency of occurrence of said reversed words in said specific language.

15. The method of claim 14 further comprising the step of:

if said input word is not identified according to the steps above, then examining said string of numbers corresponding to said input word by splitting said string of numbers corresponding to said input word into groups of n where n is an integer and equals at least 2, and comparing said groups of n with DTMF strings corresponding to snippets selected and stored according to the frequency of said snippets in said specific language.

16. The method of claim 15 in which n equals 3 and in which said string of numbers corresponding to said input word is split into at least 5 groups of 3 numbers corresponding respectively to groups including a number corresponding to the first letter of said input word, groups including a number corresponding to the second letter of the input word but excluding the number corresponding to the first letter of said input word, groups including the third letter of the input word but excluding the numbers corresponding to the first and second letters of the input word, groups including the middle letters of the input word, excluding the numbers corresponding to the first, second and third letters of the input word, and groups including the final letter of the input word.

17. The method of claim 16 further comprising the step of:

splitting said string of numbers corresponding to said input word into subgroups of 2 letters if the foregoing steps have not identified said input word and comparing said subgroups of 2 with DTMF strings corresponding to said snippets.

18. A method for determining the identity of an input word of a specific language, said method comprising the steps of:

entering said input word on a DTMF push button pad where at least one push button is identified by a number that represents more than one letter;

forming a string of numbers corresponding to said input word from said input word entered on said DTMF push button pad;

comparing said string of numbers corresponding to said input word against numerical strings corresponding to common words stored in a full word dictionary means to determine if said input word exists in said full word dictionary means;

comparing said string of numbers against numerical strings corresponding to full words and parts of words stored in a forward dictionary means wherein the letters of said full words and parts of words are stored in their natural letter sequence, and said full words and said parts of words are also stored according to the length of said full words and said parts of words and the frequency of said full words and said parts of words in said specific language; and

,

reversing said string of numbers and comparing said reversed string of numbers against numerical strings corresponding to reversed words stored in a reverse dictionary means wherein said reversed words area formed by reversing the natural letter sequence of said words and said parts of words and said reversed words are stored according to the length of said reversed words and the frequency of said reversed words in said specific language;

wherein if said input word is not located in said full word dictionary means, said string of numbers corresponding to said input word is compared against said numerical strings stored in said forward dictionary means and said string of numbers corresponding to said input word is reversed and compared against said numerical strings stored in said reverse dictionary means.

19. The method of claim 18 further comprising the step of:

comparing said string of numbers against trigrams comprising numerical representations of subgroups of 3 letters each if said input word is not identified when said string of numbers is compared against the DTMF numeric strings stored in said full word dictionary means, the DTMF numeric strings stored in said forward dictionary means or the DTMF numeric strings stored in said reverse dictionary means.

20. The method of claim 19 further comprising the step of:

comparing said string of numbers against digrams comprising numerical representations of subgroups of 2 letters each if said input word is not identified after comparing said string of numbers against the DTMF numeric strings stored in said full word dictionary means, the DTMF numeric strings stored in said forward dictionary means, the DTMF numeric strings stored in said reverse dictionary means and said trigrams.

21. The method of claim 20 wherein said trigrams are stored in at least 3 record files corresponding respectively to subgroups including Trigram I, a trigram made up of a numerical string including a number corresponding to the initial letter of a word, Trigram M, a trigram made up of a numerical string including a number corresponding to the middle of a word, and Trigram F, a trigram made up of a numerical string including a number corresponding to the last letter of a word.

22. The method of claim 21 further comprising the step of:

examining the first 3 numbers of the string of numbers corresponding to said input word and comparing them against said trigrams to determine the identity of a second letter in the input word based upon knowledge of the first letter of the input word and the frequency in said specific language of the following 2 letters based upon the knowledge of the first letter.

23. The method of claim 22 further comprising the step of:

examining the first 3 numbers of the string of numbers corresponding to said input word to determine the identity of the third letter of said input word based upon knowledge of the first 2 letters of said input word and knowledge of the frequency of the third letter of said input word in said specific language based on upon the identity of the first 2 letters of said input word.

24. The method of claim 23 further comprising the step

storing each input word after it is correctly identified;

and,

assembling said identified words into a complete message.

25. The method of claim 24 further comprising the step of:

transmitting said message in digital form over a packet transport network to receiver at a remote location; and

,

converting said digital information into speech for auditory delivery to a recipient at a remote location.

26. The method of claim 25 wherein said words stored in said full word dictionary means include proper nouns, location names, verbs, nouns, adjectives and adverbs.

27. The method of claim 26 further comprising the steps of:

storing each said input word identified by the foregoing method steps and said string of numbers corresponding to said identified input word in an internal memory dictionary; and

,

comparing subsequent number strings against said strings of numbers corresponding to said input words stored in said internal memory dictionary to determine the word identity of said number strings.

28. A method of communicating information from a sender having a telephone with DTMF buttons to a remote receiver comprising the steps of:

spelling a word letter-by-letter by utilizing the DTMF buttons on the sender'"'"'telephone;

entering a signal on one of said DTMF buttons to indicate the end of the word;

comparing each word against an electronic memory which includes a dictionary of words by determining the number of letters in said word and comparing said word against words stored in said dictionary having the same number of letters, said words also being organized according to the relative frequency for any given combination of DTMF outputs;

feeding back to the sender a guess which corresponds to the word having the highest relative frequency for a predetermined word length and for a predetermined DTMF tone combination;

repeating said comparing step if a guess is rejected and feeding back to the sender another guess at the word corresponding to the same predetermined letter length and having the next most likely relative frequency;

automatically shifting from the word guessing mode to a letter-by-letter guessing mode if the dictionary runs out of guesses for a word of a given predetermined letter length and a predetermined DTMF tone combination;

storing each word after it is correctly identified;

transmitting said word as digital data over a communication means; and

,

receiving said word at said remote receiving,

wherein words are entered by the sender and received by the receiver at a remote location.

29. An access node apparatus for sending messages from a sender having a telephone with DTMF buttons connected to a conventional telephone network over a communications means to a receiver, said access node comprising:

a telephone line interface connecting said access node to the telephone network of the sender;

a word guessing dictionary means for storing words according to their relative frequencies;

a prompter means connected to said telephone line interface and to said word guessing dictionary means for accessing said word guessing dictionary means;

speaker means connected to said prompter means and to said telephone line interface for speaking words guessed from said word guessing dictionary means to the sender;

mode shift means for automatically shifting the mode of said access node from a word guessing mode to a letter-by-letter guessing mode after the sender has exhausted all possible entries in the word guessing dictionary for a word of a given letter length and for a given DTMF tone combination sequence; and

,

message storing means connected to said prompter means for storing words prior to transmission as digital data over said communication means.