Allophonic text-to-speech generator

• US 6,148,285 A
• Filed: 10/30/1998
• Issued: 11/14/2000
• Est. Priority Date: 10/30/1998
• Status: Expired due to Term

• Current Assignee: RPX Clearinghouse LLC (RPX Corporation)
• Original Assignee: Nortel Networks Corporation
• Inventors: Busardo, Philip John
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US09/183,002
• Time in Patent Office: 746 Days
• Field of Search: 704/260, 704/258, 704/209
• 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

Abstract

The allophonic text-to-speech generator (ATTG) 10 includes a CPU 100. The CPU has a random access memory 102 and a read only memory 104 for holding the operating system, application programs, and data for the CPU 100. A keyboard 110 provides a user with control over the CPU 100. A database 130 holds phonetic transcritps of words. Such databases are well-known in the field of telephone directory assistance. A second database 140 maps allophonic text to parse and pre-recorded allophones. The CPU 100 converts a phonetic transcript of a word into an allophonic text string in accordance with a rules program 120. Then the CPU 100 extracts the audio allophone files of the allophonic string and concatenates the audio files to form the new word in the same voice as the other words fromed from the allophones in database 140.

Claims

1. A text processor for a text-to-speech synthesizer comprising:

a computer including a central processing unit having random access memory and read only memory for holding an operating system program and one or more application programs;

a phonetic text database for storing phonetic transcriptions corresponding to phonemes;

means for accessing the phonetic database to retrieve phonetic text characters corresponding to a desired word;

program means for converting the phonetic text characters into allophonic text characters to generate a string of allophonic text characters corresponding to the desired word;

an audio database comprising pre-recorded allophones stored in accordance with the allophonic text representative of each of said allophones;

means for extracting from the audio database the allophonic audio signals that correspond to the string of allophonic text in the desired word; and

means for concatenating the allophonic audio signals together to generate a new audio file corresponding to the desired word.

2. The text processor for a text-to-speech synthesizer of claim 1, further comprising an application program comprising a plurality of rules for mapping phonetic text to allophonic text.

3. The text processor for a text-to-speech synthesizer of claim 1, wherein the allophonic text file comprises a plurality of allophonic text characters for each phonetic text character.

4. A method for building speech from text with a computer including a central processing unit having random access memory and read only memory for holding an operating system program and one or more application programs, comprising the steps of:

inputting phonetic text characters corresponding to a desired spoken work;

mapping the phonetic text characters to allophonic text characters to generate a string of allophonic text characters;

providing a file of prerecorded audio signals comprising allophonic audio signals corresponding to the allophonic text characters;

extracting from the file of prerecorded audio signals the allophonic audio signals that correspond to the string of allophonic text characters; and

concatenating the allophonic audio signals together and generating an output audio signal representative of the input orthographic text.

5. The method of claim 4 wherein the audio file comprises a plurality of digital words, each word corresponding to an allophonic audio signal and the further step of converting concatenated allophonic digital audio words into analog audio signals.

6. The method of claim 4 wherein the allophonic text file comprises a plurality of allophonic text characters for each phonetic text character.