Abstract: A speech recognition text processing method includes acquiring an original text obtained by speech recognition, and performing at least one of punctuation addition processing and numerical normalization processing on the original text to obtain an intermediate result. The method further includes performing word removal processing on the original text through an attention interaction layer based on the intermediate result to obtain a processed text of the original text.

Abstract: The present disclosure provides an information switching method. The method includes: obtaining tilting information after an tilt direction of a device changes; searching a pre-set tilt direction matching the tilting information and determining pre-set information corresponding to the matched pre-set tilt direction; and switching first input information of the device to second input information, where the second input information is determined based on the pre-set information matching the pre-set tilt direction.

Abstract: Some aspects of the disclosure provide a method for image processing. The method includes receiving one or more first images corresponding to first portions in a section of characters for recognition, splicing the one or more first images into a first intermediate spliced image and performing a first intermediate character recognition on the first intermediate spliced image based on a first optical character recognition model. The first intermediate character recognition generates a first intermediate recognition result for display. The method further includes performing a final character recognition on a final spliced image corresponding to the section of characters for recognition based on a second optical character recognition model that is different from the first optical character recognition model. The final character recognition generates a final recognition result of the section of characters. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.

Abstract: Provided herein are a data processing method and apparatus, a computer device, and a storage medium for sign language translation. Sign language is translated to text by: obtaining sign language action data; determining a sign language tagging sequence corresponding to the sign language action data by element analysis, and performing operation processing on the sign language action data based on the sign language tagging sequence. Text is translated to sign language by: performing word segmentation processing on a text sequence to obtain a natural word sequence; determining a basic sign language element and an element type corresponding to each natural word in the natural word sequence; sorting the basic sign language element and the element type to generate a sign language element sequence; determining a sign language tagging sequence; and performing sign language translation processing on the text sequence based on the sign language tagging sequence.

Abstract: A handwriting recognition method is provided, which includes: obtaining handwritten original trajectory data in real-time; compressing the handwritten original trajectory data, to obtain compressed handwritten trajectory data; and inputting the compressed handwritten trajectory data into a compressed handwriting recognition model for recognition, to obtain a text recognition result corresponding to the handwritten original trajectory data. A handwriting recognition model is obtained by training with handwritten trajectory data of each piece of training data in a training data set, and the compressed handwriting recognition model is obtained by performing model compression on the handwriting recognition model. The handwriting recognition method can address the problem in the related art that the handwriting recognition accuracy is low caused by incorrect segmentation, thereby effectively improving the handwriting recognition accuracy.

Abstract: A method for poetry generation includes receiving generation information indicative of a theme for the poetry generation, and determining at least a candidate piece of poetry corresponding to the generation information according to an autoregressive language model that is configured to generate elements in the candidate piece of poetry in an autoregressive manner with a plurality of regression rounds. An element in the elements is a character or a word. The autoregressive language model is configured for generating poetry in a plurality of formats, the autoregressive language model includes a plurality of processing layers connected sequentially, a processing layer in the plurality of processing layers is configured to determine attention levels for potential elements in a potential element list according to generated elements prior to a current regression round, and predict one or more additional element for the current regression round using a neural network according to the attention levels.

Abstract: A formula recognition method and apparatus, a computer-readable medium, and an electronic device. The formula recognition method includes acquiring a target image including a formula, processing the target image to obtain a global image feature and a local image feature, and processing the global image feature and the local image feature to obtain the formula included in the target image.

Abstract: The present disclosure relates to a search method and apparatus. The method includes: obtaining network construction information corresponding to a target task, the network construction information including: search space information, sample data, and a search indicator; constructing a supernetwork based on the search space information and training the supernetwork based on the sample data, the supernetwork including a plurality of sub-networks; and searching a sub-network from the trained supernetwork based on the search indicator, to obtain a target network for performing the target task.

Abstract: An apparatus and method for data processing includes determining a search keyword according to a search statement input by a user; determining a first graph entity library corresponding to a category to which the search keyword belongs; determining, according to an association relationship between graph entity libraries of a plurality of categories, a category of a graph entity library associated with the first graph entity library, and constructing a first set, the first set including at least one graph entity library which has an association relationship with the first graph entity library, and the association relationship between the graph entity libraries is established according to a search demand of a user; and displaying, after the first set is constructed, each graph entity library included in the first set by using the search keyword as a center.

Abstract: Some aspects of the disclosure provide a method for sentence conversion. The method includes receiving a first sentence that is inputted by a user, inputting the first sentence into a first sentence based rewrite model to obtain a second sentence having a same semantic as the first sentence but a different style from the first sentence. The first sentence based rewrite model converts the first sentence into the second sentence without partitioning the first sentence into smaller portions. The method also includes displaying the second sentence.

Abstract: An apparatus and method for entry recommendation includes: recalling a plurality of vocabularies based on a multi-path recall policy, entries in each of the vocabularies having original scores; normalizing the original scores for the vocabulary to obtain target scores of the entries in the vocabulary; respectively determining weights of the entries in the vocabulary; selecting target entries from the plurality of vocabularies based on the target scores and the weights; and pushing the target entries to a target client.

Abstract: An input method, an input device, and an apparatus for input are provided in the embodiments of the present application. The method specifically includes: receiving an input string having a fast input intent, wherein the fast input intent is used to indicate, according to a shorthand information of a word or a phrase corresponding to the input string, the word or the phrase; obtaining word candidates and/or phrase candidates corresponding to the input string according to a language model, wherein the word candidates and the phrase candidates are respectively complete words and complete phrases corresponding to the input string; presenting word candidates and/or phrase candidates to a user. The embodiments of the present application can not only improve the flexibility and application range of the fast input, but also improve the quality of word candidates and/or phrase candidates, thereby improving input efficiency.

Abstract: In present disclosure, a data processing method, a data processing device, and an apparatus for data processing are provided. The method specifically includes: receiving a source language speech input by a target user; determining, based on the source language speech, a target acoustic model from a preset acoustic model library, the acoustic model library including at least two acoustic models corresponding to different timbre characteristics; converting, based on the target acoustic model, the source language speech into a target language speech; and outputting the target language speech. According to the embodiments of the present disclosure, the recognition degree of the speaker corresponding to the target language speech output by the translation device can be increased, and the effect of user communication can be improved.

Abstract: The present disclosure provides a recommendation method and device, and a device for formulating recommendations. The recommendation method specifically includes obtaining a current context of an input method program after the input method program is triggered, analyzing data of past input actions by a current user under past contexts to obtain a target entry matching the current context in the data of the past input actions, and displaying the target entry. The present disclosure provides candidate entries to the user without any input operation by the user, thereby substantially improving the user's input efficiency.

Abstract: An information processing method and device for simplifying the operation process. The method comprises: in a process of performing interaction via a pre-configured messaging mode, acquiring interaction information sent by an opposite end; determining a preset triggering condition that matches the interaction information through analysis, and acquiring corresponding display information; and recommending the display information corresponding to the matched preset triggering condition. The information that the user may be interested in during the interaction process is directly recommended and displayed, such that the operation process is simplified, and the efficiency is improved.

Abstract: Embodiments of the present disclosure provide a text editing method, device, and electronic apparatus. The method comprises: acquiring a text-to-be-edited; after receiving a trigger operation, generating an edit region; segmenting the text-to-be-edited into N number of text segments, and separately displaying the N number of text segments in the edit region, where N is a positive integer greater than 1; and after receiving an editing operation with respect to the N number of text segments, editing the N number of text segments based on the editing operation. The method and device provided in the embodiments of the present disclosure are used to solve technical problems such as complicated operations and low editing efficiency in current text editing technologies. The embodiments of the present disclosure may reduce complexity of text editing operations and improve technical effects of text editing efficiency.

Abstract: The present disclosure provides a method for adjusting an input virtual keyboard and an input apparatus, which are directed to solve the problem of one-hand operation in existing input virtual keyboards. The method comprises: based on a received keyboard floating instruction: adjusting a size of an input window to a screen size; intercepting a click event performed on a touchscreen by a user and detecting the click event; and when the click event is detected as a keyboard adjustment instruction, based on the keyboard adjustment instruction, adjusting at least one of a drawing position and a drawing size of an input view, wherein the input view is a region for drawing the input virtual keyboard in the input window.

Abstract: An audio input method includes: in an audio-input mode, receiving a first audio input by a user, recognizing the first audio to generate a first recognition result, and displaying corresponding verbal content to the user based on the first recognition result; and in an editing mode, receiving a second audio input by the user and recognizing and generating a second recognition result, converting the second recognition result to an editing instruction, and executing a corresponding operation based on the editing operation. The audio-input mode and the editing mode are switchable.

Abstract: The present disclosure provides a voice collection device, comprising: a housing; a microphone assembly disposed inside the housing and including a microphone main body and a microphone circuit board; and a sealing structure disposed between the microphone assembly and an inner side of the housing and including a dust filter and a first adhesive layer. One side of the microphone circuit board is connected to the dust filter. The other side of the microphone circuit board is connected to the microphone main body. The microphone circuit board and the dust filter are bonded to the inner side of the housing. A sound hole is configured on the microphone main body. A first through-hole is configured at a position on the microphone circuit board corresponding to the sound hole. A second through-hole is configured at a position on the housing corresponding to the first through-hole.