Abstract: A speech recognition device includes: an acquisition part, acquiring a speech signal; a speech feature amount calculation part, calculating a speech feature amount; a first speech recognition part, based on the speech feature amount, performing speech recognition using a learned first E2E model, attaching a first tag to a vocabulary portion of a specific class in text that is a recognition result, and outputting the same; a second speech recognition part, based on the speech feature amount, performing speech recognition using a learned second E2E model, attaching a second tag to a vocabulary portion of a specific class in a phoneme that is a recognition result, and outputting the same; a phoneme replacement part, replacing a vocabulary with the first tag with a phoneme with the second tag; and an output part, converting the phoneme with the second tag into text and outputting the same.

Abstract: The speech recognition that is disclosed analyzes an acoustic feature for each subframe of an audio signal; provides a first model configured to determine a hidden state for each frame consisting of multiple subframes on the basis of the acoustic feature; provides a second model configured to determine a hidden state for each frame consisting of multiple subframes on the basis of the acoustic feature; and provides a third model configured to determine an utterance content on the basis of a sequence of the hidden states of each block consisting of multiple frames belonging to a voice segment.

Abstract: A speech recognition device includes an acquisition unit configured to acquire audio data of an utterance and a speech recognition unit configured to generate text from the audio data using an automatic speech recognition model. The automatic speech recognition model includes an audio encoder configured to convert the audio data into a feature, a bias encoder configured to convert a registered bias token into a feature, and a bias decoder expanded to correspond to a bias token and configured to estimate the next token on the basis of a feature output by the audio encoder, a feature output by the bias encoder, and a previously estimated token sequence.

Abstract: A voice recognition device includes an acquisition unit which acquires a frame per unit time of a voice stream, a streaming feature generation unit which generates a first feature from the frame using a streaming encoder, a streaming character generation unit which generates a first character from the first feature using a streaming decoder, a non-streaming feature generation unit which generates a second feature sequence from a first feature sequence obtained by joining the first feature of each of the plurality of frames using a non-streaming encoder, a streaming character generation unit which generates a second character string from the second feature sequence using a plurality of non-streaming decoders, and a learning unit which performs Knowledge Distillation between the streaming encoder and the non-streaming encoder on the basis of the first feature sequence and the second feature sequence.

Abstract: A conversation support device includes: an acquisition unit configured to acquire speech; a display unit configured to display text based on the acquired speech; and a control unit configured to cause the display unit to display the text based on the speech. The control unit recognizes a piece of text selected by a user out of a plurality of pieces of the text displayed on the display unit as designated text. The control unit causes the display unit to display speech associated with the designated text in correlation with the designated text when the speech associated with the designated text is input in a state in which the designated text has been recognized.

Abstract: A voice display device includes: an acquisition unit configured to acquire a voice signal; a voice recognizing unit configured to recognize the acquired voice signal; a display unit configured to display a voice recognition result; and a processing unit configured to start an application for displaying the voice recognition result of the voice signal including a wake word when the wake word is acquired and to cause the display unit to display the voice recognition result of the voice signal including the wake word.

Abstract: This paper proposes an attention-based contextual biasing method that can be customized using an editable phrase list (referred to as a bias list). The proposed method can be trained effectively by combining a bias phrase index loss and special tokens to detect the bias phrases in the input speech data.

Abstract: A speech recognition device is a speech recognition device for performing speech recognition using an end-to-end model and includes an encoder that is a model that converts an input speech signal into a characteristic expression, a decoder that is a model that converts speech data into text using an output of the encoder, and a learning unit that randomly selects a block length of the speech signal input to the encoder and causes the encoder and the decoder to learn.

Abstract: A speech recognition device includes: an acquisition part, acquiring a speech signal; a speech feature amount calculation part, calculating a speech feature amount; a first speech recognition part, based on the speech feature amount, performing speech recognition using a learned first E2E model, attaching a first tag to a vocabulary portion of a specific class in text that is a recognition result, and outputting the same; a second speech recognition part, based on the speech feature amount, performing speech recognition using a learned second E2E model, attaching a second tag to a vocabulary portion of a specific class in a phoneme that is a recognition result, and outputting the same; a phoneme replacement part, replacing a vocabulary with the first tag with a phoneme with the second tag; and an output part, converting the phoneme with the second tag into text and outputting the same.

Abstract: The speech recognition that is disclosed analyzes an acoustic feature for each subframe of an audio signal; provides a first model configured to determine a hidden state for each frame consisting of multiple subframes on the basis of the acoustic feature; provides a second model configured to determine a hidden state for each frame consisting of multiple subframes on the basis of the acoustic feature; and provides a third model configured to determine an utterance content on the basis of a sequence of the hidden states of each block consisting of multiple frames belonging to a voice segment.

Abstract: In detecting abnormal sound of a test object such as a vehicle immediately after completion of vehicle assembly, a reference sound data is recorded as reference data having phase and amplitude in each of multiple frequency bands reference sounds of types not previously recorded as abnormal sounds in the test object similar to the test object, test sound data is acquired which has phase and amplitude in the multiple frequency bands from test sounds generated by the test object. And sound feature data is acquires by comparing the test sound data with reference sound data in frequency bands the same as the multiple frequency bands and by changing amplitude of the test sound data based on amplitude of the reference sound data while maintaining phase of the test sound data, and abnormal sound is detected based on the acquired sound feature data.

Abstract: In determining presence/absence of abnormal sound occurrence in a test object such as a vehicle immediately after completion of vehicle assembly, sound data of the test object during running on a rough test track is recorded and resolved into multiple frequency bands by time unit. The sound data resolved into multiple frequency bands by time unit is compared among the multiple frequency bands and correlation coefficients matrix values indicating strength of correlation between the multiple frequency bands n are calculated. Presence/absence of the abnormal sound occurrence is finally determined based on the calculated correlation coefficients matrix values.

Abstract: In detecting abnormal sound of a test object such as a vehicle immediately after completion of vehicle assembly, a reference sound data is recorded as reference data having phase and amplitude in each of multiple frequency bands reference sounds of types not previously recorded as abnormal sounds in the test object similar to the test object, test sound data is acquired which has phase and amplitude in the multiple frequency bands from test sounds generated by the test object. And sound feature data is acquires by comparing the test sound data with reference sound data in frequency bands the same as the multiple frequency bands and by changing amplitude of the test sound data based on amplitude of the reference sound data while maintaining phase of the test sound data, and abnormal sound is detected based on the acquired sound feature data.

Abstract: In determining presence/absence of abnormal sound occurrence in a test object such as a vehicle immediately after completion of vehicle assembly, sound data of the test object during running on a rough test track is recorded and resolved into multiple frequency bands by time unit. The sound data resolved into multiple frequency bands by time unit is compared among the multiple frequency bands and correlation coefficients matrix values indicating strength of correlation between the multiple frequency bands n are calculated. Presence/absence of the abnormal sound occurrence is finally determined based on the calculated correlation coefficients matrix values.