Data processing: speech signal processing – linguistics – language – Speech signal processing – Synthesis
Reexamination Certificate
2000-03-14
2001-06-05
Dorvil, Richemond (Department: 2741)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Synthesis
C704S258000, C704S003000, C704S002000, C455S412100, C709S217000, C709S206000, C709S207000
Reexamination Certificate
active
06243681
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a speech synthesizer for converting text data to speech data and outputting the data, and particularly to a speech synthesizer that can be used in CTI (Computer Telephony Integration) systems.
2. Description of the Related Art
In recent years, speech synthesizers for artificially making and outputting speech using digital signal processing techniques have become widespread. In particular, in CTI systems that implement a phone handling service providing a high degree of customer satisfaction integrating computer systems and telephone systems, use of a speech synthesizer makes it possible to provide the contents of electronic mail etc. transferred across a computer network as speech output through a telephone on the public network.
A speech output service (called a unified message service hereafter) in such a CTI system is implemented as described in the following. For example, when voice output is carried out for electronic mail, a CTI server constituting the CTI system co-operates with a mail server responsible for the electronic mail, and in response to a call arrival signal from a telephone on the public network, electronic mail at an address indicated at the time of the call arrival signal is acquired from the mail server, and at the same time text data contained in that electronic mail is converted to speech data using a speech synthesizer installed in the CTI server. By transmitting the speech data after conversion to the telephone of the caller, the CTI server allows the user of that telephone to begin listening to the contents of the electronic mail. In providing a unified message service, for example, the CTI server cooperates with a WWW (world wide web) server, so that the WWW server can turn some (portions made up of sentences) of content (for example a web page) submitted on a computer network such as the internet into speech output.
A speech synthesizer of the related art, particularly a speech synthesizer installed in a CTI server, is usually made to cope specifically with one particular language, for example Japanese. On the other hand, items to be converted, such as electronic mail etc. exist in various languages such as Japanese and English.
Accordingly, with the speech synthesizer of the related art, it was not really possible to correctly carry out conversion to speech data by matching the language supported by the speech synthesizer with the language of text data to be converted. For example, if an English sentence is converted using a speech synthesizer that supports Japanese, the sentence structures are different in Japanese and English with respect to syntax, grammar etc., which means that compared to when conversion is carried out using a speech synthesizer supporting English, it was difficult to provide high quality speech output because correct speech output was not possible and speech output was not fluent.
Particularly in the CTI system, in the case where speech output is carried out using the unified message service, high quality speech output can not be carried out because the telephone subscriber judges the content of electronic mail etc. only from results of speech output, with the result that erroneous contents may be conveyed.
SUMMARY OF THE INVENTION
The object of the present invention is to provide a speech synthesizer that can perform high quality speech output, even when text data to be converted is in various languages.
In order to achieve the above described object, a speech synthesizer of the present invention is provided with a plurality of voice synthesizing means for converting text data to speech data, with each speech synthesizing means converting text data in different languages to speech data in languages corresponding to those of the text data, wherein conversion of specific text data to speech data is selectively carried out by one of the plurality of speech synthesizing means.
With the above described speech synthesizer, a plurality of speech synthesizing means supporting respectively different languages are provided, and one of the plurality of speech synthesizing means selectively carries out conversion from text data to speech data. Accordingly, by using this speech synthesizer it is possible to carry out conversion to speech data even if text data in various languages are to be converted, by using the speech synthesizing means supporting each language.
REFERENCES:
patent: 4829580 (1989-05-01), Church
patent: 5412712 (1995-05-01), Jennings
patent: 5615301 (1997-03-01), Rivers
patent: 5991711 (1999-11-01), Seno et al.
patent: 6085162 (2000-07-01), Cherny
Systranet™ (Systran Translation Technologies) advertisement, Jul. 2000.
Guji Yoshiki
Ohtsuki Koji
Dorvil Richemond
Frank Robert J.
Gluck Jeffrey W.
Nolan Daniel A.
OKI Electric Industry Co., Ltd.
LandOfFree
Multiple language speech synthesizer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multiple language speech synthesizer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiple language speech synthesizer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2500189