30 results about "Speech reading" patented technology

There is disclosed a method and system for preparing a document to be read by a text-to-speech reader. The method can include identifying two or more voice types available to the text-to-speech reader, identifying the text elements within the document, grouping related text elements together, and classifying the text elements according to voice types available to the text-to-speech reader. The method of grouping the related text elements together can include syntactic and intelligent clustering. The classification of text elements can include performing latent semantic analysis on the text elements and characteristics of the available voice types.

A system and method for using semantic analysis to configure a voice reader is presented. A text file includes a plurality of text blocks, such as paragraphs. Processing performs semantic analysis on each text block in order to match the text block's semantic content with a semantic identifier. Once processing matches a semantic identifier with the text block, processing retrieves voice attributes that correspond to the semantic identifier (i.e. pitch value, loudness value, and pace value) and provides the voice attributes to a voice reader. The voice reader uses the text block to produce a synthesized voice signal with properties that correspond to the voice attributes. The text block may include semantic tags whereby processing performs latent semantic indexing on the semantic tags in order to match semantic identifiers to the semantic tags.

An Internet accessed server that on demand downloads and activates text-to-speech program elements to a subscriber's computer. Program elements are customized to match the operating system of the subscriber's computer. Upon termination of the text-to-speech session, the server deactivates all program elements and the subscriber becomes free to reinitiate over the Internet the text-to-speech program on the same or another computer system.

A digital broadcast receiving device, characterized in that it includes a receiving and selecting unit (3) for receiving and selecting digital broadcast waves, and for decoding data broadcast streams included in the received and selected digital broadcast waves , and obtain a transport stream decoder and a data broadcast decoder (4, 8) for character data and image data for data broadcasting, a character recognition dictionary memory (22) for storing patterns and reading data of the patterns, for An image recognition unit (23) that compares the obtained image data with information stored in the character recognition dictionary memory (22), and obtains reading data of patterns contained in the image data, for The obtained character data and the acquired reading data generate a reading sentence generating unit (24) for sentence data of phonetic reading, and a speech synthesis unit for converting the generated sentence data into phonetic data (25).

The invention discloses a segment segmentation detection method and system, the method comprising: obtaining the segment segmentation result obtained by segmenting the reading voice; performing voice recognition on the reading voice to obtain the recognition result; according to the The reading text corresponding to the reading voice calculates the recognition rate of the segmentation result and the recognition rate of the recognition result respectively; if the recognition rate of the segmentation result is higher than the recognition rate of the recognition result, then determine that the segmentation result is normal; otherwise It is determined that the segmentation result is abnormal. The invention also discloses a spoken language evaluation method and system. The present invention can effectively detect whether the segment segmentation result is abnormal, thereby ensuring the accuracy of the evaluation result.

An ultrasonic thickness gauge with voice reading function is used in the field of non-destructive testing. It includes two parts, the host and the probe. The probe part is the ultrasonic pulse transmitting unit and the receiving unit. The host part includes a CPU, an ultrasonic transmitting circuit, a signal receiving and amplifying circuit, a thickness wave forming circuit, a memory circuit, a counter circuit, a display circuit and a keyboard. , memory circuit is used for storing thickness value, is characterized in that: also be provided with the voice circuit that is connected with CPU, USB universal serial bus interface circuit; USB interface circuit is made up of the USB interface integrated circuit with parallel bus and CPU connection, voice circuit It is used to realize the voice reporting; the CPU controls the ultrasonic thickness measuring circuit, and the thickness value of the measured object is obtained through calculation and processing, which can be displayed, and the measurement data can also be read out by the CPU controlling the voice circuit. All measurement data are finally uploaded to the PC through the USB interface for data analysis and collection. The instrument is practical.

The invention provides an electronic reading apparatus and a voice reading loading method. The voice reading loading method includes the following steps: a typesetting engine conducting type setting on an analyzed text, and the type setting includes: segmenting the text in according with preset sentence segmenting rules and recording in successive order coordinate information of each sentence and a corresponding sentence number; conducting sentence segmenting on the text which has been subject to the type setting in accordance with the preset sentence segmenting rules and adding a only identifier at the end of each sentence that has been segmented so as to form a new sentence with the only identifier; sending the newly segmented sentence to a voice synthesizing engine so as to form audio data and recording the position of a starting character corresponding with the new sentence and also recording the number of the synthesized sentences; executing voice reading from the selected sentence.

The invention discloses a voice map reading method and system based on a smart TV. The method comprises the following steps: receiving a picture and character information corresponding to the picture sent by a front end device; displaying the picture on the smart TV, and obtaining the corresponding character information according to the picture; and converting the character information into voice information, and synchronously playing the character information with the picture. By adoption of the voice map reading method and system disclosed by the invention, the smart TV increases a picture voice reading function, the character information corresponding to the picture is converted into the voice information, and the voice information is synchronously played with the picture by human voice, thereby being convenient for users to read the character information corresponding to the picture and bringing convenience to the use of the users.

The invention discloses an end-to-end intelligent voice reading evaluation method, which includes: collecting the data processing flow of reading voice, target pronunciation, error code, and speaker information used for training and evaluating the neural network; Auxiliary training; according to the target pronunciation to be evaluated and the pronunciation to be evaluated, the evaluation results are directly output from the neural network end-to-end, from the input of the target pronunciation and the pronunciation to be evaluated to the output of the evaluation results, the entire process can be differentiated, and can be directly targeted at the evaluation indicators optimization. The present invention directly constructs an end-to-end evaluation mode in which the input is speech and text to be evaluated, and the output is feedback results. It can be jointly trained with each module in the method to make the overall effect better. In addition, the auxiliary tasks constructed by the method can more accurately extract Evaluate related features to make feedback evaluation results more accurate.