Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1998-03-27
2001-04-17
Dorvil, Richemond (Department: 2641)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S270000, C704S235000
Reexamination Certificate
active
06219644
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to the field of dictation interfaces for speech applications, and in particular, to an audio interface for facilitating dictation when no visual interface is available.
2. Description of Related Art
As people begin to use computer speech-to-text dictation applications more and more, one of the most common places for dictation will be in environments where their eyes and hands are otherwise busy. By definition, these users will be unable to receive any visual feedback in an eyes-busy environment. One of the most effective dictation techniques is for the user to first write out a list of points or bullets that the user needs to cover when the user dictates. However, in an eyes-busy, hands-busy environment it can be unwieldy or sometimes even impossible to use a written list of bullets, even if prepared beforehand, as for example when driving a car.
Experience with dictation in these environments suggests that some sort of non-visual prompting mechanism needs to be developed. These environments are extremely important for support because they will be the most useful environments for notes and letters. Given the accuracy rates of today's dictation engines, even under ideal editing conditions, a good typist will always outperform someone dictating. However, when typing is difficult or impossible, dictation can be an extremely productive alternative.
When dictating without a visible interface, the user is unable to tell what template field the user is dictating into, and likely unable to remember all of the fields the user needs to dictate into, for example when dictating into a form.
SUMMARY OF THE INVENTION
The solution to this problem is to provide a list of points or bullets to the person dictating. Because the visual environment is occupied, the next best free input modality is the auditory modality. Therefore, the user can be accommodated by audio prompts that give the user information about the field or section the user is about to dictate into. The audio prompts can be programmed into a document template, in which each field of the document is provided with an audio point or bullet which can be played back as each field or section is sequentially accessed. To the extent the user does remember any of the field or section names, such fields or sections can be accessed directly by a command, for example Go To <fieldname>. The template can also be programmed with an initial audio prompt which identifies the template as well as a final audio prompt which advises the user that the template has been completed and can be saved, if not earlier named.
A method for enabling dictation into a speech application from an audio-only interface, in accordance with an inventive arrangement, can include the following steps beginning with generating a visual user interface for creating a dictation template. The template preferably has a plurality of named dictation fields and respective audio prompts identifying each of the dictation fields by a respective name. Second, the template can be opened in response to a command spoken through the audio-only interface. Third, a first one of the audio prompts corresponding to a first one of the dictation fields can be transmitted through the audio-only interface. Fourth, dictation can be accepted into the first one of the dictation fields through the audio-only interface. Fifth, a subsequent one of the dictation fields can be opened in response to another command spoken through the audio-only interface. Sixth, a subsequent one of the audio prompts corresponding to the subsequent one of the dictation fields can be transmitted through the audio-only interface. Seventh, dictation can be accepted into the subsequent one of the dictation fields through the audio-only interface. Finally, each of the fifth, sixth and seventh steps can be repeated until the dictation is complete. In this way, a user can dictate into the dictation fields without seeing or remembering the names of the dictation fields.
The dictation fields are advantageously accessible in a predetermined sequence without being identified by name in the spoken commands. The dictation fields are also advantageously accessible when identified by name in the spoken commands.
An audio template for enabling dictation into a speech application from an audio-only interface, in accordance with another inventive arrangement, can include a plurality of named dictation fields and respective audio prompts identifying each of the dictation fields by a respective name. The audio template can be openable in response to a command spoken through the audio-only interface. Moreover, each of the dictation fields can be openable in response to a spoken command transmitted through the audio-only interface. Finally, each of the respective audio prompts can be automatically transmittable through the audio-only interface in response to the dictation fields being opened. In this way, a user can dictate into the dictation fields without seeing or remembering the names of the dictation fields.
The dictation fields are advantageously accessible in a predetermined sequence without being identified by name in the spoken commands. The dictation fields are also advantageously accessible when identified by name in the spoken commands.
REFERENCES:
patent: 5265075 (1993-11-01), Bergeron et al.
patent: 5477511 (1995-12-01), Englehardt
patent: 5875448 (1999-02-01), Boys et al.
patent: 5884262 (1999-09-01), Wise et al.
patent: 5950167 (1999-09-01), Yaker
Stifelman, LJ et al., “Voicenotes: A Speech Interface for a Hand-Held Voice Notetaker,” Conference Proceedings on Human Factors in Computing Systems, pp. 179-18, 1993.*
Schindler, E., “The Computer Speech Book”, Academic Press, pp. 221-249, 1996.
Armstrong Angelo
Dorvil Richemond
International Business Machines Corp.
Senterfitt Akerman
LandOfFree
Audio-only user speech interface with audio template does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Audio-only user speech interface with audio template, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Audio-only user speech interface with audio template will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2496442