Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1998-08-03
2001-05-01
Dorvil, Richemond (Department: 2741)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S275000
Reexamination Certificate
active
06226615
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to a method of displaying text to be spoken by a speaker, and to apparatus for such display. Such a system can be used in a television studio to generate a television signal.
When a person is delivering a speech from a previously prepared script, for example a newsreader in a televised news broadcast or someone addressing a conference audience, a visual prompting system is frequently used. This presents the words to the speaker within their natural line of sight by using partially-reflecting mirrors appropriately positioned, for example, across the lens of a television camera.
FIG. 1
shows a perspective view of a typical arrangement currently used in a television studio whilst
FIG. 2
is a side view showing how the partially-reflecting mirror allows the speaker to read the projected text of the script whilst looking into the lens of the television camera.
In the system
100
of
FIGS. 1 and 2
, a newsreader
112
is seated before a television camera
114
. The television camera is provided with a partially-reflecting mirror
116
over the camera lens, inclined at 45° to the horizontal and leaning away from the camera towards the newsreader. Below the partially-reflecting mirror
116
is positioned a text display device
118
, for example a computer monitor in the form of a cathode ray tube display device. Text displayed on the display device
118
can be seen by the newsreader
112
as, being in the camera lens, it is in the line of sight of the newsreader. Just a few words can be displayed in this way;
FIG. 1
shows the display of the words “The incident was seen”. The display is driven by a suitable computer, for example based on a standard personal computer.
This arrangement allows the presenter to see the next few words of the script. An operator controls the scrolling of the projected text smoothly upwards to match the spoken delivery made by the presenter. In this way the presenter is able to read from the text without looking away from the camera.
Whilst this works very effectively, it requires an operator to control the scrolling of the text in time with the speaker.
International Patent Application WO-A-90/01203 describes a system using speech recognition in conjunction with a display, as part of a language training system to train a student to speak a foreign language. The quality of the speech is indicated to the user, and then the system moves on to new text. In International Patent Application WO-A-94/10666 is described a children's speech training aid which compares a child's speech with stored models of speech, and gives an indication of whether or not the child has spoken correctly. Finally, European Patent Application EP-A-0 649 144 describes a method of automatically indexing recorded audio using speech recognition, for use in aligning text stored in computer files and a recording. The system using indexing is complicated and would be difficult to operate in real time.
SUMMARY OF THE INVENTION
The invention in its various aspects is defined in the independent claims below. Advantageous features are set forth in the appendant claims.
A preferred embodiment of the invention is described in more detail below with reference to the drawings. In this embodiment, text to be spoken by a speaker, e.g. a newsreader, and held in a store, is displayed on a display which is seen by the speaker through a partially-reflecting mirror mounted in front of a television camera, which is operating to produce a live television signal for broadcasting. The text is scrolled up as the speaker speaks, this being achieved automatically by a controller. The controller recognises the spoken speech picked up by a microphone, using speech recognition techniques, and matches the spoken text with the stored text to be spoken. The speed at which the text is spoken is thus determined and used by the controller to adjust the rate at which the text scrolls up on the display. The system can be used in live broadcasts.
Preferably, if the speech recognition fails, the text continues to be changed at a rate dependent upon the previous rate of change.
REFERENCES:
patent: 3972603 (1976-08-01), Lubinec
patent: 4472833 (1984-09-01), Turrell et al.
patent: 4894722 (1990-01-01), Suzuki
patent: 5477282 (1995-12-01), Moore
patent: 5799273 (1998-08-01), Mitchell et al.
patent: 6064961 (2000-05-01), Hanson
patent: 0 262 938 (1988-04-01), None
patent: 0649144 A1 (1995-04-01), None
patent: 0 649 144 (1995-04-01), None
patent: 08-056296 (1996-02-01), None
patent: 09065176 (1997-03-01), None
patent: WO 90/01203 (1990-02-01), None
patent: WO 94/10666 (1994-05-01), None
Global Telecommunication Conference 1988. Globecom '88. Bazzani et al., “PC-Based telephone communication system for deaf-blind people”. Pp. 43-47 vol. 1. Dec. 1988.*
“Open-Vocabulary Speech Indexing for Voice and Video Mail Retrieval”, M.G. Brown et al, ACM Multimedia 96, Boston MA USA, pp. 307-331.
“A Fast Lattice-Based Approach to Vocabulary Independent Wordspotting”, D.A. James and S.J. Young, Proceedings of the International Conference on Acoustics, Speed and Signal Processing, ICASSP 94, vol. 1, pp. 377-380, Adelaide, IEEE.
Dormand Neil
Kirby David Graham
British Broadcasting Corporation
Dick & Harris
Dorvil Richemond
LandOfFree
Spoken text display method and apparatus, for use in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Spoken text display method and apparatus, for use in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Spoken text display method and apparatus, for use in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2447126