Industrial Technology Research Institute

Shih-Chieh Chien

Sen-Chia Chang

Chih-Chung Kuo

Yao-Yuan Chang

Jia-Jang Tu

Chung-Jen Chiu

Hsin-Chang Chang

Chung-Mou Penwu

Jui-Hsin Hung

A speech recognition system and method thereof, a vocabulary establishing method and a computer program product are provided. The speech recognition method includes: storing a speech recognition model including speech-units and basic components of acoustic models, wherein each of the speech-units includes at least one state and each state corresponds to one of the basic components of acoustic models; receiving first and second speech signals; obtaining a speech-unit sequence of a native/non-native vocabulary from a speech-analysis and unit-expansion module; recognizing the first speech signal according to the speech recognition model and the speech-unit sequence of the native/non-native vocabulary and further outputting a recognition result; and selecting an optimal component from the basic components of acoustic models according to the speech recognition model, the second speech signal, and the word corresponding to the second speech signal, and further updating the speech-units according to the best basic component of acoustic model.

Speech recognition system and method thereof, vocabulary establishing method and computer program product

In a spoken word generation system for speech recognition, at least one input device receives a plurality of input signals at least including at least one sound signal; a mode detection module detects the plurality of input signals; when a specific sound event is detected in the at least one sound signal or at least one control signal is included in the plurality of input signals, a speech training mode is outputted; when no specific sound event is detected in the at least one sound signal and no control signal is included in the plurality of input signals, a speech recognition mode is outputted; a speech training module receives the speech training mode and performs a training process on the audio segment and outputs a training result; and a speech recognition module receives the speech recognition mode, and performs a speech recognition process and outputs a recognition result.

Spoken word generation method and system for speech recognition and computer readable medium thereof

A method and apparatus for utterance verification are provided for verifying a recognized vocabulary output from speech recognition. The apparatus for utterance verification includes a reference score accumulator, a verification score generator and a decision device. A log-likelihood score obtained from speech recognition is processed by taking a logarithm of the value of the probability of one of feature vectors of an input speech conditioned on one of states of each model vocabulary. A verification score is generated based on the processed result. The verification score is compared with a predetermined threshold value so as to reject or accept the recognized vocabulary.

Method and apparatus for utterance verification

A system for leaving and transmitting speech messages automatically analyzes input speech of at least a reminder, fetches a plurality of tag informations, and transmits speech message to at least a message receiver, according to the transmit criterions of the reminder. A command or message parser parses the tag informations at least including at least a reminder ID, at least a transmitted command and at least a speech message. The tag informations are sent to a message composer for being synthesized into a transmitted message. A transmitting controller controls a device switch according to the reminder ID and the transmitted command, to allow the transmitted message send to the message receiver via a transmitting device.

System and method for leaving and transmitting speech messages

The present invention relates to a speech interactive system and method. The system comprises a target information receiving module, an interactive mode setting and speech processing module, an interactive information update module, a decision module, and an output response module. It receives target information and sets corresponding target text sentence information. It also receives a user's speech signal, sets an interactive mode, decides the speech's target text sentence information, and generates an assessment for the target text sentence. Under the set interactive mode, the system updates the information in an interactive information recording table according to the assessment and a timing count. According to the interactive mode and the recorded information, an output mode for the target text sentence information is generated. According to the output mode and the recorded information, the response information is generated.

Speech interactive system and method

A method and a system for executing correlative services are provided. In the method and the system, an event type corresponding to an input message is determined through semantic analysis. After collecting the necessary execution information of the event type according to the input message, a user database, or by inquiring the user or another system, the system automatically executes various correlative services of the event type. Therefore, the system can help users to execute correlative services more correctly and more efficiently.

Method and system for executing correlative services

A method and system for utterance verification is disclosed. It first extracts a sequence of feature vectors from speech signal. At least one candidate string is obtained after speech recognition. Then, speech signal is segmented into speech segments according to the verification-unit-specified structure of candidate string for making each speech segment corresponding to a verification unit. After calculating the verification feature vectors of speech segments, these verification feature vectors are sequentially used to generate verification scores of speech segments in verification process. This invention uses neural networks for calculating verification scores, where each neural network is a Multi-Layer Perceptron (MLP) developed for each verification unit. Verification score is obtained through using feed-forward process of MLP. Finally, utterance verification score is obtained by combining all verification scores of speech segments and is used to compare with a pre-defined threshold for the decision of acceptance or rejection of the candidate string.

Method and system for utterance verification

A method for generating candidate word strings in speech recognition is provided, which is based on the nodes in the word lattice to search candidate word strings. The associated maximum string score for each node is first determined. Next, all nodes are sorted based on the associated maximum string score to group the nodes with the same string score into the same node set. Then, the node sets with relative high string scores are selected to connect the nodes by their starting time frame and ending time frame, thereby generating the candidate word strings.

Method for generating candidate word strings in speech recognition

In speaker-independent speech recognition, between-speaker variability is one of the major resources of recognition errors. A speaker cluster model is used to manage recognition problems caused by between-speaker variability. In the training phase, the score function is used as a discriminative function. The parameters of at least two cluster-dependent models are adjusted through a discriminative training method to improve performance of the speech recognition.

Speech recognition method using speaker cluster models

Device and method of channel effect compensation for a telephone speech recognition system is disclosed. The telephone speech recognition system comprises a compensatory neutral network and a recognize. The compensatory neural network receives an input signal and compensates the input signal with a bias to generate an output signal. The compensatory neural network provides a plurality of first parameters to determine the bias. The recognizer is coupled to the compensatory neural network for classifying the output signal according to a plurality of second parameters in acoustic models to generate a recognition result and determine a recognition loss. The first parameters and second parameters are adjusted according to the recognition loss and an adjustment means during a training process.

Growing community of inventors

Taichung, Taiwan

Shih-Chieh Chien