Device for generating announcement information

• US 5,621,891 A
• Filed: 11/19/1992
• Issued: 04/15/1997
• Est. Priority Date: 11/19/1991
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: US Philips Corporation (Koninklijke Philips N.V.)
• Inventors: Meyer, Peter, Ruhl, Hans-Wilhelm
• Primary Examiner(s): Hafiz, Tariq R.
• Application Number:
  US07/978,097
• Time in Patent Office: 1,608 Days
• Field of Search: 395/2, 395/2.1, 395/2.14-2.15, 395/2.21, 395/2.3, 395/2.65-2.68, 395/2.75-2.77, 395/2.86-2.87, 381/51, 381/77, 379/88
• US Class Current: 704/270
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers

Abstract

The invention relates to a device for generating announcement information. When the complete announcement information is generated via natural speech information, a large storage capacity is required. The device aims to enable a plurality of different announcement information to be generated without requiring a large storage capacity. To this end, there is proposed a device for generating announcement information including an input unit, a storage unit for storing natural speech information, a speech generator for generating synthetic speech information and a multiplexer for combining the natural speech information and the synthetic speech information to form the announcement information.

Claims

1. A device for generating speech information signal, comprising:

input means for generating a first control signal and a second control signal that are mutually exclusive in time;

storage means for storing natural speech information in the form of first time-sequential strings each having one or more distinct words;

speech generating means for, under control of said second control signal, generating synthetic speech information as further time-sequential strings each having one or more further distinct words; and

multiplexing means coupled to said storage means and to said speech generating means for receiving said natural speech information and said synthetic speech information and for selectively outputting on a word-by-word basis one of said natural speech information and said synthetic speech information in response to said first control signal and said second control signal, respectively, to generate said speech information signal, while being blocked from switching-over between said storage means and said speech generating means other than at the end of said time-sequential string.

2. The device as claimed in claim 1, wherein said natural speech information includes a speech block and wherein said synthetic speech information includes a key word.

3. The device as claimed in claim 2, wherein said speech information signal includes a sentence including a plurality of speech blocks and a key word interposed between two of said speech blocks.

4. The device as claimed in claim 1, wherein said natural speech information is stored in said storage means in encoded form and wherein said synthetic speech information generated by said speech generating means is encoded in conformity with said code of said natural speech information.

5. The device as claimed in claim 4, wherein said storage means stores frequency variation information for a junction between one of said speech blocks and an adjacent one of said key words.

6. The device as claimed in claim 5, wherein said speech generating means responsive to said frequency variation information, changes a parameter of said synthetic speech information.

7. The device as claimed in claim 1, wherein said storage means stores frequency variation information for a junction between one of said speech blocks and an adjacent one of said key words.

8. The device as claimed in claim 7, wherein said speech generating means responsive to said frequency variation information, changes a parameter of said synthetic speech information.

9. The device as claimed in claim 8, wherein said output means is responsive to said input means.

10. The device as claimed in claim 1, further including output means coupled to said multiplexing means for outputting said speech information signal, said output means including an output memory and a digital-to-analog converter.

11. The device as claimed in claim 10, wherein said speech generating means includes a speech model based on speech data from said one speaker.

12. The device as claimed in claim 1, wherein said natural speech information is derived from one speaker.

13. The device as claimed in claim 1, further including a microphone coupled to said input means for receiving said natural speech information.