A system and method are disclosed for retrieving audio segments from a spoken document. The spoken document preferably is one having moderate word error rates such as telephone calls or teleconferences. The method comprises converting speech associated with a spoken document into a lattice representation and indexing the lattice representation of speech. These steps are performed typically off-line. Upon receiving a query from a user, the method further comprises searching the indexed lattice representation of speech and returning retrieved audio segments from the spoken document that match the user query.

A method of conducting computing experiments, includes executing a set of jobs, performing a comparison of a result of the executed set of jobs with templates of previously-executed experiments which are stored in a knowledge base, and identifying a prunable job of the set of jobs based on the comparison and a user constraint.

A semiconductor integrated circuit device for speech recognition includes a conversion candidate setting unit that receives text data indicating words or sentences together with a command and sets the text data in a conversion list in accordance with the command; a standard pattern extracting unit that extracts, from a speech recognition database, a standard pattern corresponding to at least a part of the words or sentences indicated by the text data that is set in the conversion list; a signal processing unit that extracts frequency components of an input speech signal and generates a feature pattern indicating distribution of the frequency components; and a match detecting unit that detects a match between the feature pattern generated from at least a part of the speech signal and the standard pattern and outputs a speech recognition result.

Generally, this disclosure provides systems, devices, methods and computer readable media for adaptation of language models and semantic tracking to improve automatic speech recognition (ASR). A system for recognizing phrases of speech from a conversation may include an ASR circuit configured to transcribe a user's speech to a first estimated text sequence, based on a generalized language model. The system may also include a language model matching circuit configured to analyze the first estimated text sequence to determine a context and to select a personalized language model (PLM), from a plurality of PLMs, based on that context. The ASR circuit may further be configured to re-transcribe the speech based on the selected PLM to generate a lattice of paths of estimated text sequences, wherein each of the paths of estimated text sequences comprise one or more words and an acoustic score associated with each of the words.

A method for spoken term detection, comprising generating a time-marked word list, wherein the time-marked word list is an output of an automatic speech recognition system, generating an index from the time-marked word list, wherein generating the index comprises creating a word loop weighted finite state transducer for each utterance, i, receiving a plurality of keyword queries, and searching the index for a plurality of keyword hits.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for voice recognition. In one aspect, a method includes the actions of receiving a voice input; determining a transcription for the voice input, wherein determining the transcription for the voice input includes, for a plurality of segments of the voice input: obtaining a first candidate transcription for a first segment of the voice input; determining one or more contexts associated with the first candidate transcription; adjusting a respective weight for each of the one or more contexts; and determining a second candidate transcription for a second segment of the voice input based in part on the adjusted weights; and providing the transcription of the plurality of segments of the voice input for output.

A method for spoken term detection, comprising generating a time-marked word list, wherein the time-marked word list is an output of an automatic speech recognition system, generating an index from the time-marked word list, wherein generating the index comprises creating a word loop weighted finite state transducer for each utterance, i, receiving a plurality of keyword queries, and searching the index for a plurality of keyword hits.

In some embodiments, the recognition results produced by a speech processing system (which may include a top recognition result and one or more alternative recognition results) based on an analysis of a speech input, are evaluated for indications of potential significant errors. In some embodiments, the recognition results may be evaluated to determine whether a meaning of any of the alternative recognition results differs from a meaning of the top recognition result in a manner that is significant for the domain. In some embodiments, one or more of the recognition results may be evaluated to determine whether the result(s) include one or more words or phrases that, when included in a result, would change a meaning of the result in a manner that would be significant for the domain.

The present invention is directed generally to linking a collection of words and/or phrases with locations in a video and/or audio stream where the words and/or phrases occur and/or associations of a collection of words and/or phrases with a call history.

Systems and processes for converting speech-to-text are provided. In one example process, speech input can be received. A sequence of states and arcs of a weighted finite state transducer (WFST) can be traversed. A negating finite state transducer (FST) can be traversed. A virtual FST can be composed using a neural network language model and based on the sequence of states and arcs of the WFST. The one or more virtual states of the virtual FST can be traversed to determine a probability of a candidate word given one or more history candidate words. Text corresponding to the speech input can be determined based on the probability of the candidate word given the one or more history candidate words. An output can be provided based on the text corresponding to the speech input.