Method for dynamically altering text-to-speech (TTS) attributes of a TTS engine not inherently capable of dynamic attribute alteration

• US 6,188,983 B1
• Filed: 09/02/1998
• Issued: 02/13/2001
• Est. Priority Date: 09/02/1998
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Hanson, Gary Robert
• Primary Examiner(s): Zele, Krista
• Assistant Examiner(s): Opsasnick, Michael N.
• Application Number:
  US09/145,781
• Time in Patent Office: 895 Days
• Field of Search: 704/260, 704/270
• US Class Current: 704/260
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• 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

In a TTS system external to a TTS engine, a method for dynamically altering text-to-speech (TTS) attributes of a TTS engine not inherently capable of dynamic attribute alteration is disclosed. The method of the invention can include several steps beginning with the step of representing in the external system a resettable start word position value by a first variable. Second, a current word position offset value can be represented in the external system by a second variable. Third, the second variable can be updated in the external system each time a word of the specified text is played back. Fourth, the TTS playback can be stopped in response to a user request to alter the TTS attribute. Fifth, after stopping the TTS playback, the TTS engine attribute can be altered. Also, after stopping the TTS playback, the first variable can be replaced in the external system with a new start word position value corresponding to the next successive unplayed word of the specified text. Finally, playback of the text can be restarted in accordance with the new word start position value to play back the next successive unplayed word.

Claims

1. In a system external to a TTS engine, a method for dynamically altering text-to-speech (TTS) playback attributes of the TTS engine during playback of specified text in accordance with at least one TTS engine attribute, wherein the TTS engine is not inherently capable of dynamic attribute alteration, comprising the steps of:

representing in said external system a resettable start word position value by a first variable, said resettable start word position indicating a starting offset of said specified text currently loaded in said TTS engine;

representing in said external system a current word position value by a second variable, said current word position value indicating an offset relative to said resettable start word position;

updating in said external system said second variable each time a word of said specified text is played back;

stopping said TTS playback of said TTS engine in response to a user request to alter said at least one TTS attribute;

after stopping said TTS playback, altering said at least one TTS engine attribute;

also after stopping said TTS playback, replacing in said external system said first variable with a new start word position value corresponding to the next successive unplayed word of said specified text; and

,

restarting playback of said text in accordance with said new word start position value, said next successive unplayed word being played back by said TTS engine.

2. The method of claim 1, further comprising the step of resetting said second variable to zero after replacing said first variable with said new start word position value corresponding to said next successive unplayed word of said specified text.

3. The method of claim 2, wherein said step of replacing said first variable comprises the steps of:

replacing said first variable with a sum of said first and second variables; and

,

using said replaced first variable to determine said new start word position value.

4. The method of claim 1, wherein said step of replacing said first variable comprises the steps of:

replacing said first variable with a sum of said first and second variables; and

,

using said replaced first variable to determine said new start word position value.

5. The method of claim 1, wherein after said stopping step, said altering step begins prior to said replacing step.

6. The method of claim 1, wherein after said stopping step, said replacing step begins prior to said altering step.

7. The method of claim 1, wherein after said stopping step, said altering step and said replacing step are time coincident.

8. The method of claim 1, comprising the step of stopping said TTS engine playback only between words of said specified text.

9. The method of claim 1, further comprising the step of representing a current word position in said specified text by a sum of said first and second variables.

10. The method of claim 1, further comprising the steps of:

receiving said user request in a TTS client;

transmitting a request to stop said playback from said TTS client to said TTS engine;

transmitting a notice from said TTS engine to said TTS client that said playback has stopped;

transmitting a request to alter said at least one TTS engine attribute from said TTS client to said TTS engine;

transmitting a notice from said TTS engine to said TTS client that said at least one TTS engine attribute has been altered;

sending a text string from said TTS client to said TTS engine beginning with said next successive unplayed word; and

,

playing back said text string beginning with said next successive unplayed word.