Unified messaging system with voice messaging and text messaging using text-to-speech conversion

• US 6,477,494 B2
• Filed: 01/07/2000
• Issued: 11/05/2002
• Est. Priority Date: 07/03/1997
• Status: Expired due to Term

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Hyde-Thomson, Henry C. A., Liron, Roger
• Primary Examiner(s): Banks-Harold, Marsha D.
• Assistant Examiner(s): LERNER, MARTIN
• Application Number:
  US09/479,333
• Publication Number:
  US 20020069062A1
• Time in Patent Office: 1,033 Days
• Field of Search: 704/260, 704/270, 704/277, 704/258, 704/266, 704/269, 379/88.05, 379/88.06, 379/88.13, 379/88.14, 379/88.16
• US Class Current: 704/260
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/005 Language recognition
• CPC Class Codes: H04L 51/066 Format adaptation, e.g. format conversion or compression
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 3/5307 for recording messages comprising any combination of audio and non-audio components

Abstract

A unified messaging system selectively provides voice messages and computer-generated speech messages to a set of subscribers. The unified messaging system includes a voice gateway server coupled to an electronics mail system and a private branch exchange (PBX). In addition, the unified messaging system is configured for determining if a received message is a text message from the electronic mail system or a voice message from the PBX. The voice gateway server preferably includes a trigraph analyzer. When a subscriber selects review of a received message that includes a text message, the trigraph analyzer sequentially examines 3-character combinations within the text message; determines occurrence frequencies for the character combinations; compares the occurrence frequencies with reference occurrence statistics modeled from text samples written in particular languages; and generates a language identifier and a likelihood value for the text message. Based upon the language identifier, a message inquiry unit selects an appropriate text-to-speech engine for converting the text message into computer-generated speech that is played to the subscriber.

Claims

1. A unified messaging system for selectively providing voice messages from received voice messages and received text messages, the voice messaging system comprising:

a. means for retrieving a received message;

b. means for determining if the received message is an original text message or an original voice message, wherein the means for determining is coupled to the means for retrieving the received message;

c. means for automatically identifying a language of the original text message, wherein the means for automatically identifying is coupled to the means for determining;

d. means for converting the original text message into a computer-generated speech message wherein the means for converting includes a plurality of text-to-speech engines wherein each of the plurality of text-to-speech engines corresponds to one of a plurality of languages and the means for converting selects an appropriate one text-to-speech engine from the plurality of text-to-speech engines based on the identified language of the original text message, wherein the means for converting is coupled to the means for retrieving the received message; and

e. means for playing a selected one of the original voice message and the computer-generated speech message to a subscriber, wherein the means for playing is coupled to the means for retrieving and the means for converting and the selected one of the original voice message and the computer-generated speech message is selected by a message recipient.

2. A method of receiving messages, the method comprising:

a. retrieving a current message;

b. determining a message format of the current message as one of a voice format and a text format;

c. generating a converted speech message based on the current message when the message format of the current message is the text format, wherein generating the converted speech message includes automatically identifying a language of the current message, selecting an appropriate one text-to-speech engine from a plurality of text-to-speech engines wherein each of the plurality of text-to-speech engines corresponds to one of a plurality of languages and the appropriate one text-to-speech engine is selected based upon the language identified and converting the text format of the current message into the converted speech message based on the language of the current message;

d. playing the voice format of the current message to a message recipient in response to a message play back command from the message recipient, when the message format of the current message is the voice format; and

e. playing the converted speech message to the message recipient in response to a message play back command from the message recipient, when the message format of the current message is the text format.

3. The method as claimed in claim 2, wherein the step of automatically identifying the language of the current message further comprises determining a frequency occurrence for each of a plurality of sequential character combinations within the text format of the current message, comparing the frequency occurrence for each of the plurality of sequential character combinations with reference occurrence statistics for each of a plurality of reference languages and determining the language of the current message based on a best match of the frequency occurrence for each of the plurality of sequential character combinations and one of the reference occurrence statistics.

4. The method as claimed in claim 3, wherein the step of determining the frequency occurrence for each of the plurality of sequential character combinations further comprises examining at least 175 characters within the text format of the current message.

5. The method as claimed in claim 3, wherein each of the plurality of sequential character combinations further comprises three characters.