Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.

Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.

Abstract: Optimized language models are provided for in-domain applications through an iterative, joint-modeling approach that interpolates a language model (LM) from a number of component LMs according to interpolation weights optimized for a target domain. The component LMs may include class-based LMs, and the interpolation may be context-specific or context-independent. Through iterative processes, the component LMs may be interpolated and used to express training material as alternative representations or parses of tokens. Posterior probabilities may be determined for these parses and used for determining new (or updated) interpolation weights for the LM components, such that a combination or interpolation of component LMs is further optimized for the domain. The component LMs may be merged, according to the optimized weights, into a single, combined LM, for deployment in an application scenario.

Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.

Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.

Abstract: Query history expansion may be provided. Upon receiving a spoken query from a user, an adapted language model may be applied to convert the spoken query to text. The adapted language model may comprise a plurality of queries interpolated from the user's previous queries and queries associated with other users. The spoken query may be executed and the results of the spoken query may be provided to the user.

Abstract: Systems and methods are provided for optimizing language models for in-domain applications through an iterative, joint-modeling approach that expresses training material as alternative representations of higher-level tokens, such as named entities and carrier phrases. From a first language model, an in-domain training corpus may be represented as a set of alternative parses of tokens. Statistical information determined from these parsed representations may be used to produce a second (or updated) language model, which is further optimized for the domain. The second language model may be used to determine another alternative parsed representation of the corpus for a next iteration, and the statistical information determined from this representation may be used to produce a third (or further updated) language model. Through each iteration, a language model may be determined that is further optimized for the domain.

Abstract: The customization of recognition of speech utilizing context-specific language model scale factors is provided. Training audio may be received from a source in a training phase. The received training audio may be recognized utilizing acoustic and language models being combined utilizing static scale factors. A comparison may then be made of the recognition results to a transcription of the training audio. The recognition results may include one or more hypotheses for recognizing speech. Context specific scale factors may then be generated based on the comparison. The context specific scale factors may then be applied for use in the speech recognition of audio signals in an application phase.

Abstract: Query history expansion may be provided. Upon receiving a spoken query from a user, an adapted language model may be applied to convert the spoken query to text. The adapted language model may comprise a plurality of queries interpolated from the user's previous queries and queries associated with other users. The spoken query may be executed and the results of the spoken query may be provided to the user.

Abstract: The customization of language modeling components for speech recognition is provided. A list of language modeling components may be made available by a computing device. A hint may then be sent to a recognition service provider for combining the multiple language modeling components from the list. The hint may be based on a number of different domains. A customized combination of the language modeling components based on the hint may then be received from the recognition service provider.

Abstract: An incremental speech recognition system. The incremental speech recognition system incrementally decodes a spoken utterance using an additional utterance decoder only when the additional utterance decoder is likely to add significant benefit to the combined result. The available utterance decoders are ordered in a series based on accuracy, performance, diversity, and other factors. A recognition management engine coordinates decoding of the spoken utterance by the series of utterance decoders, combines the decoded utterances, and determines whether additional processing is likely to significantly improve the recognition result. If so, the recognition management engine engages the next utterance decoder and the cycle continues. If the accuracy cannot be significantly improved, the result is accepted and decoding stops.

Abstract: Various components provide options to re-format an input based on one or more contexts. The input is received that has been submitted to an application (e.g., messaging application, mobile application, word-processing application, web browser, search tool, etc.), and one or more outputs are identified that are possibilities to be provided as options for re-formatting. A respective score of each output is determined by applying a statistical model to a respective combination of the input and each output, the respective score comprising a plurality of context scores that quantify a plurality of contexts of the respective combination. Exemplary contexts include historical-user contexts, domain contexts, and general contexts. One or more suggested outputs are selected from among the one or more outputs based on the respective scores and are provided as options to re-format the input.

Abstract: Query history expansion may be provided. Upon receiving a spoken query from a user, an adapted language model may be applied to convert the spoken query to text. The adapted language model may comprise a plurality of queries interpolated from the user's previous queries and queries associated with other users. The spoken query may be executed and the results of the spoken query may be provided to the user.

Abstract: A semantic error rate calculation may be provided. After receiving a spoken query from a user, the spoken query may be converted to text according to a first speech recognition hypothesis. A plurality of results associated with the converted query may be received and compared to a second plurality of results associated with the converted query.

Abstract: On a computing device a speech utterance is received from a user. The speech utterance is a section of a speech dialog that includes a plurality of speech utterances. One or more features from the speech utterance are identified. Each identified feature from the speech utterance is a specific characteristic of the speech utterance. One or more features from the speech dialog are identified. Each identified feature from the speech dialog is associated with one or more events in the speech dialog. The one or more events occur prior to the speech utterance. One or more identified features from the speech utterance and one or more identified features from the speech dialog are used to calculate a confidence score for the speech utterance.

Abstract: Architecture that employs an overall grammar as a set of context-specific grammars for recognition of an input, each responsible for a specific context, such as subtask category, geographic region, etc. The grammars together cover the entire domain. Moreover, multiple recognitions can be run in parallel against the same input, where each recognition uses one or more of the context-specific grammars. The multiple intermediate recognition results from the different recognizer-grammars are reconciled by running re-recognition using a dynamically composed grammar based on the multiple recognition results and potentially other domain knowledge, or selecting the winner using a statistical classifier operating on classification features extracted from the multiple recognition results and other domain knowledge.

Abstract: A received utterance is recognized using different language models. For example, recognition of the utterance is independently performed using a baseline language model (BLM) and using an adapted language model (ALM). A determination is made as to what results from the different language model are more likely to be accurate. Different features may be used to assist in making the determination (e.g. language model scores, recognition confidences, acoustic model scores, quality measurements, . . . ) may be used. A classifier may be trained and then used in determining whether to select the results using the BLM or to select the results using the ALM. A language model may be automatically trained or re-trained that adjusts a weight of the training data used in training the model in response to differences between the two results obtained from applying the different language models.

Abstract: A Statistical Machine Translation (SMT) model is trained using pairs of sentences that include content obtained from one or more content sources (e.g. feed(s)) with corresponding queries that have been used to access the content. A query click graph may be used to assist in determining candidate pairs for the SMT training data. All/portion of the candidate pairs may be used to train the SMT model. After training the SMT model using the SMT training data, the SMT model is applied to content to determine predicted queries that may be used to search for the content. The predicted queries are used to train a language model, such as a query language model. The query language model may be interpolated other language models, such as a background language model, as well as a feed language model trained using the content used in determining the predicted queries.

Abstract: On a computing device a speech utterance is received from a user. The speech utterance is a section of a speech dialog that includes a plurality of speech utterances. One or more features from the speech utterance are identified. Each identified feature from the speech utterance is a specific characteristic of the speech utterance. One or more features from the speech dialog are identified. Each identified feature from the speech dialog is associated with one or more events in the speech dialog. The one or more events occur prior to the speech utterance. One or more identified features from the speech utterance and one or more identified features from the speech dialog are used to calculate a confidence score for the speech utterance.

Abstract: A semantic error rate calculation may be provided. After receiving a spoken query from a user, the spoken query may be converted to text according to a first speech recognition hypothesis. A plurality of results associated with the converted query may be received and compared to a second plurality of results associated with the converted query.