System for generating speech and non-speech audio messages

• US 6,760,704 B1
• Filed: 09/29/2000
• Issued: 07/06/2004
• Est. Priority Date: 09/29/2000
• Status: Expired due to Term

• Current Assignee: Beijing Xiaomi Mobile Software Co., Ltd. (Xiaomi Corp.)
• Original Assignee: Intel Corporation
• Inventors: Bennett, Steven M.
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US09/676,104
• Time in Patent Office: 1,376 Days
• Field of Search: 455/412, 455/413, 704/270, 704/275, 379/88.08, 379/88.12, 379/88.16, 379/88.18, 379/88.22
• US Class Current: 704/270
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems

Abstract

An audio information system that may be used to form and convey an audio message having speech overlapped with non-speech audio is provided. The system has components to store a context indicator having non-speech audio to signify a characteristic of a speech content stream, to merge the context indicator with the speech content stream to form an integrated message, and to output the integrated message. The message has overlapping non-speech audio from the context indicator and speech audio. The system also has mechanisms to vary the format of integrated message generated in order to train the user on non-speech cues. In addition, other aspects of the present invention relating to the audio information system receiving content and generating an audio message are described.

Claims

1. An audio information system comprising:

a storage unit to store a context indicator having non-speech audio to signify a characteristic of a speech content stream;

a combination unit to merge the context indicator with the speech content stream to form an integrated message having the non-speech audio of the context indicator overlapped with speech audio;

an outlet port to output the integrated message; and

a tracking unit to determine user training of the context indicator.

2. The audio information system of claim 1, wherein the merging is by overlapping the context indicator with at least a portion of the speech content stream.

3. The audio information system of claim 1, wherein the context indicator contains a speech audio overlapped with the non-speech audio.

4. The audio information system of claim 1, wherein the determination is by counting the number of times that the integrated message having the characteristic is outputted to a user over a given time period and comparing the outputted number to a predetermined number.

5. The audio information system of claim 1, wherein the context indicator includes a speech description of the characteristic if the user requires training.

6. The audio information system of claim 1, wherein the storage unit is to store more than one different type of speech content stream and the context indicator signifies the type of speech content stream to merge.

7. The audio information system of claim 1, wherein the storage unit is to store more than one speech content stream and the system further comprising a selection unit to select the speech content stream from the storage unit to merge with the context indicator.

8. The audio information system of claim 7, wherein the selecting of the speech information stream is based on a priority determination.

9. A method for generating an audio message, comprising:

storing a context indicator having non-speech audio to signify a characteristic of a speech content stream;

merging the context indicator with the speech content stream to form an integrated message having the non-speech audio of the context indicator overlapped with speech audio;

outputting the integrated message; and

determining user training of the context indicator.

10. The method of claim 9, wherein the merging is by overlapping the context indicator with at least a portion of the speech content stream.

11. The method of claim 9, wherein the context indicator contains a speech audio overlapped with the non-speech audio.

12. The method of claim 9, wherein the determining is by counting the output number of times that the integrated message having the characteristic is outputted to a user over a given time period and comparing the outputted number to a predetermined number.

13. The method of claim 9, wherein the context indicator includes a speech description of the characteristic if the user requires training.

14. The method of claim 9, further including determining the characteristic of the speech content stream.

15. The method of claim 9, wherein more than one different type of speech content stream is stored and the context indicator signifies the type of speech content stream to merge.

16. The method of claim 9, wherein more than one speech content stream is stored and the method further includes selecting the speech content stream from the storage unit to merge with the context indicator.

17. The method of claim 16, wherein the selecting of the speech information stream is based on a priority determination.

18. A computer readable medium having stored therein a plurality of sequences of executable instructions, which, when executed by an audio information system for generating an audio message, cause the system to:

store a context indicator having non-speech audio to signify a characteristic of a speech content stream;

merge the context indicator with the speech content stream to form an integrated message having the non-speech audio of the context indicator overlapped with speech audio;

output the integrated message; and

to determine user training on the context indicator.

19. The computer readable medium of claim 18, wherein the merging is by overlapping the context indicator with at least a portion of the speech content stream.

20. The computer readable medium of claim 18, wherein the context indicator contains a speech audio overlapped with the non-speech audio.

21. The computer readable medium of claim 18, wherein the determination is by counting the number of times that the integrated message having the characteristic is outputted to a user over a given time period and comparing the outputted number to a predetermined number.

22. The computer readable medium of claim 18, wherein the context indicator includes a speech description of the characteristic if the user requires training.

23. The computer readable medium of claim 18, further including additional sequences of executable instructions, which, when executed by the audio information system, cause the system to determine the characteristic of the speech content stream.

24. The computer readable medium of claim 18, wherein the characteristic is speech content type, speech content source, or relevance of the speech content stream.

25. The computer readable medium of claim 18, wherein more than one different type of speech content stream is stored and the context indicator signifies the type of speech content stream to merge.

26. The computer readable medium of claim 18, wherein more than one speech content stream is stored and further including selecting the speech content stream from the storage unit to merge with the context indicator.

27. The computer readable medium of claim 26, wherein the selecting of the speech information stream is based on a priority determination.