摘要:
Automated methods and apparatus for synchronizing audio and text data, e.g., in the form of electronic files, representing audio and text expressions of the same work or information are described. Also described are automated methods of detecting errors and other discrepancies between the audio and text versions of the same work. A speech recognition operation is performed on the audio data initially using a speaker independent acoustic model. The recognized text in addition to audio time stamps are produced by the speech recognition operation. The recognized text is compared to the text in text data to identify correctly recognized words. The acoustic model is then retrained using the correctly recognized text and corresponding audio segments from the audio data transforming the initial acoustic model into a speaker trained acoustic model. The retrained acoustic model is then used to perform an additional speech recognition operation on the audio data. The audio and text data are synchronized using the results of the updated acoustic model. In addition, one or more error reports based on the final recognition results are generated showing discrepancies between the recognized words and the words included in the text. By retraining the acoustic model in the above described manner, improved accuracy is achieved.
摘要:
Automated methods and apparatus for synchronizing audio and text data, e.g., in the form of electronic files, representing audio and text expressions of the same work or information are described. A statistical language model is generated from the text data. A speech recognition operation is then performed on the audio data using the generated language model and a speaker independent acoustic model. Silence is modeled as a word which can be recognized. The speech recognition operation produces a time indexed set of recognized words some of which may be silence. The recognized words are globally aligned with the words in the text data. Recognized periods of silence, which correspond to expected periods of silence, and are adjoined by one or more correctly recognized words are identified as points where the text and audio files should be synchronized, e.g., by the insertion of bi-directional pointers. In one embodiment, for a text location to be identified for synchronization purposes, both words which bracket, e.g., precede and follow, the recognized silence must be correctly identified. Pointers, corresponding to identified locations of silence to be used for synchronization purposes are inserted into the text and/or audio files at the identified locations. Audio time stamps obtained from the speech recognition operation may be used as the bi-directional pointers. Synchronized text and audio data may be output in a variety of file formats.
摘要:
The transmission of information during ad click-through is disclosed. In one embodiment, a computer-implemented method selects an ad to be displayed on a web page, as one of a plurality of ads within a current cluster in which each of the ad has a probability to be selected. The method displays the ad on the web page, and then detects activation—for example, click-through—of the displayed ad. The method transmits information to an entity associated with the ad, such as an advertiser, upon detecting click-through or other activation of the ad. In one embodiment, the information transmitted includes information regarding the current cluster.
摘要:
Targeted delivery of items with inventory management using a cluster-based approach or a rule-based approach is disclosed. An example of items is advertisements. Each item is allocated to one or more clusters. The allocation is made based on a predetermined criterion accounting for at least a quota for each item and possibly a constraint for each cluster. The former can refer to the number of times an item must be shown. The latter can refer to the number of times a given group of web pages is likely to be visited by users, and hence is the number of times items can be shown in a given cluster. The invention is not limited to any particular definition of what constitutes a cluster or item.
摘要:
High-quality composite printing elements are prepared without the need for precise registration of constituent photocurable elements by disposing at least one photocurable element, and preferably a plurality of photocurable elements, upon a surface of a printing element in approximate register and then transferring a computer-generated negative onto a surface of the elements.
摘要:
An electronic metronome device producing precisely timed and tuned rhythms and pitches that are pre-programmed to correspond to specific scales or modes, arpeggios, chords, and etudes. A combination of microprocessor and user interface (34, 30, 28, and 26) stores these musical exercises and retrieves them from an electronic memory (36), inputs them to a signal processor (42) for amplification and modification, and outputs (40) them to speakers (12), optical displays (24), audio outputs (18), etc. A volume control 20 and balance control 22 modify the audio signal coming from the speakers 12. The components, enabled either by an internal (battery) or external (plug) power source (44), are housed in a light and durable case for easy portability and user control.
摘要:
A modeling method for predicting a decision is disclosed. A risk environment is simulated for one or more control groups. One or more experimental groups are exposed to an intervention, and the risk environment is simulated for the experimental groups.
摘要:
High-quality composite printing elements are prepared without the need for precise registration of constituent photocurable elements by disposing at least one photocurable element, and preferably a plurality of photocurable elements, upon a surface of a printing element in approximate register and then transferring a computer-generated negative onto a surface of the elements.
摘要:
An online commerce system facilitates online commerce over a public network using an online commerce card. The "card" does not exist in physical form, but instead exists in digital form. It is assigned a customer account number that includes digits for a prefix number for bank-handling information, digits for a customer identification number, digits reserved for an embedded code number, and a digit for check sum. The bank also gives the customer a private key. During an online transaction, the customer computer retrieves the private key and customer account number from storage. The customer computer generates a code number as a function of the private key, customer-specific data (e.g, card-holder's name, account number, etc.) and transaction-specific data (e.g., transaction amount, merchant ID, goods ID, time, transaction date, etc.). The customer computer embeds the code number in the reserved digits of the customer account number to create a transaction number specific to the transaction. The customer submits that transaction number to the merchant as a proxy for a regular card number. When the merchant submits the number for approval, the issuing institution recognizes it as a proxy transaction number, indexes the customer account record, and looks up the associated private key and customer-specific data. The institution computes a test code number using the same function and input parameters as the customer computer. The issuing institution compares the test code number with the code number embedded in the transaction number. If the two numbers match, the issuing institution accepts the transaction number as valid.
摘要:
There is disclosed a method and apparatus for subtitling stereoscopic imagery. The stereoscopic imagery may include a plurality of paired stereo images having a perspective and providing a stereoscopic scene, wherein each image of a given pair represents a perspective of the imagery as viewed by a single eye of the stereoscopic scene. A subtitle may be presented solely upon one image of the stereo pair of images of at least some of the stereoscopic imagery.