Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 8
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

INGENIOUS: the first responder of the future. A next generation integrated toolkit for collaborative response, increasing protection and augmenting operational capacity.

  • Institute of Communications and Computer Systems
  • Universiteit Twente
  • Deutsches Zentrum Fuer Luft – Und Raumfahrt
  • Diginext
  • Fundacion Tekniker
  • Exus
  • SINTEF
  • Center for Research & Technology Hellas
  • Totalförvarets Forskningsinstitut
  • SATWAYS
  • Alpes Lasers
  • Technische Universitaet Wien
  • Cyprus Research & Innovation Center
  • Universitat Pompeu Fabra
  • SingularLogic
  • Korea Institute of Robot & Convergence
  • Gobierno Vasco, Departamento de Seguridad
  • Hopitaux de Paris
Comunicació presentada a: SafeCorfu 2019, the 6th International Conference on Civil Protection & New Technologies, celebrat a Corfu, Grècia, del 6 al 9 de novembre de 2019., The INGENIOUS project (EU Horizon 2020) aims to develop, integrate, test, deploy, demonstrate and validate a Next Generation Integrated Toolkit (NGIT) for Collaborative Response, which ensures high level of Protection & Augmented Operational Capacity to respond to the disaster scene. This will comprise a multitude of the tools and services required: 1) for enabling protection of the first responders with respect to their health, safety and security; 2) for enhancing their operational capacities by offering them with means to conduct various response tasks and missions boosted with autonomy, automation, precise positioning, optimal utilisation of available resources and upgraded awareness and sense‐making; 3) for allowing shared response across first responders teams and disciplines by augmenting their field of view, information sharing and communications between teams and with victims. The NGIT armours the FRs at all fronts, by delivering novel, affordable, accepted and customised response tools and services as part of their uniform and as part of their operational assets. The NGIT will be provided at the service of the FRs for extensive testing and validation (at component and Toolkit levels) in the framework of a rich Training, Testing and Validation Programme – of Lab Tests (LSTs), Small‐Scale Field Tests (SSTs) and Full‐Scale Field Validations (FSXs) – towards powering the FR of the future being fully aware, fully connected and fully integrated., This work is supported by the projects INGENIOUS, which has received funding from the European Union's Horizon 2020 (H2020) programme under grant agreement No:833435.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Naturalness enhancement with linguistic information in end-to-end TTS using unsupervised parallel encoding

  • Peiró Lilja, Àlex
  • Farrús, Mireia
Comunicació presentada a Interspeech 2020 celebrat del 25 al 29 d'octubre de 2020 a Shanghai, Xina., State-of-the-art end-to-end speech synthesis models have reached levels of quality close to human capabilities. However, there is still room for improvement in terms of naturalness, related to prosody, which is essential for human-machine interaction. Therefore, part of current research has shift its focus on improving this aspect with many solutions, which mainly involve prosody adaptability or control. In this work, we explored a way to include linguistic features into the sequenceto- sequence Tacotron2 system to improve the naturalness of the generated voice. That is, making the prosody of the synthesis looking more like the real human speaker. Specifically we embedded with an additional encoder part-of-speech tags and punctuation mark locations of the input text to condition Tacotron2 generation. We propose two different architectures for this parallel encoder: one based on a stack of convolutional plus recurrent layers, and another formed by a stack of bidirectional recurrent plus linear layers. To evaluate the similarity between real read-speech and synthesis, we carried out an objective test using signal processing metrics and a perceptual test. The presented results show that we achieved an improvement in naturalness., This work is a part of the INGENIOUS project, funded by the European Union’s Horizon 2020 Research and Innovation Programme and the Korean Government under Grant Agreement No 833435. The second author has been funded by the Agencia Estatal de Investigaci´on (AEI), Ministerio de Ciencia, Innovaci ´on y Universidades and the Fondo Social Europeo (FSE) under grant RYC-2015-17239 (AEI/FSE, UE). This work has been carried out using an NVIDIA GPU Titan Xp generously provided by NVIDIA Company.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Data augmentation for low-resource Quechua ASR improvement

  • Zevallos, Rodolfo
  • Bel Rafecas, Núria
  • Cámbara Ruiz, Guillermo
  • Farrús, Mireia
  • Luque, Jordi
Comunicació presentada a INTERSPEECH 2022, celebrat del 18 al 22 de setembre de 2022 a Inchon, Corea del Sud., Automatic Speech Recognition (ASR) is a key element in new services that helps users to interact with an automated system. Deep learning methods have made it possible to deploy systems with word error rates below 5% for ASR of English. However, the use of these methods is only available for languages with hundreds or thousands of hours of audio and their corresponding transcriptions. For the so-called low-resource languages to speed up the availability of resources that can improve the performance of their ASR systems, methods of creating new resources on the basis of existing ones are being investigated. In this paper we describe our data augmentation approach to improve the results of ASR models for low-resource and agglutinative languages. We carry out experiments developing an ASR for Quechua using the wav2letter++ model. We reduced WER by 8.73% through our approach to the base model. The resulting ASR model obtained 22.75% WER and was trained with 99 hours of original resources and 99 hours of synthetic data obtained with a combination of text augmentation and synthetic speech generation., This work has been partially supported by the Project PID2019-104512GB-I00, Ministerio de Ciencia e Innovación and Agencia Estatal de Investigación (Spain), and the INGENIOUS project from the European Union’s Horizon 2020 Research and Innovation Program under grant numbers 833435. The third author has been funded by the Agencia Estatal de Investigación (AEI), Ministerio de Ciencia, Innovación y Universidades and the Fondo Social Europeo (FSE) under grant RYC-2015-17239 (AEI/FSE, UE).
Dipòsit Digital de la UB
Publicación Artículo científico (JournalArticle)

TASE: Task-Aware Speech Enhancement for Wake-Up Word Detection in Voice Assistants

  • Cámbara, Guillermo
  • López, Fernando
  • Bonet, David
  • Gómez, Pablo
  • Segura, Carlos
  • Farrús, Mireia
  • Luque, Jordi
Wake-up word spotting in noisy environments is a critical task for an excellent user experience with voice assistants. Unwanted activation of the device is often due to the presence of noises coming from background conversations, TVs, or other domestic appliances. In this work, we propose the use of a speech enhancement convolutional autoencoder, coupled with on-device keyword spotting, aimed at improving the trigger word detection in noisy environments. The end-to-end system learns by optimizing a linear combination of losses: a reconstruction-based loss, both at the log-mel spectrogram and at the waveform level, as well as a specific task loss that accounts for the cross-entropy error reported along the keyword spotting detection. We experiment with several neural network classifiers and report that deeply coupling the speech enhancement together with a wake-up word detector, e.g., by jointly training them, significantly improves the performance in the noisiest conditions. Additionally, we introduce a new publicly available speech database recorded for the Telefónica's voice assistant, Aura. The OK Aura Wake-up Word Dataset incorporates rich metadata, such as speaker demographics or room conditions, and comprises hard negative examples that were studiously selected to present different levels of phonetic similarity with respect to the trigger words 'OK Aura'. Keywords: speech enhancement; wake-up word; keyword spotting; deep learning; convolutional neural network
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

TASE: Task-Aware Speech Enhancement for Wake-Up Word Detection in Voice Assistants

  • Cámbara, Guillermo
  • López, Fernando
  • Bonet, David
  • Gómez, Pablo
  • Segura, Carlos
  • Farrús, Mireia
  • Luque, Jordi
Wake-up word spotting in noisy environments is a critical task for an excellent user experience with voice assistants. Unwanted activation of the device is often due to the presence of noises coming from background conversations, TVs, or other domestic appliances. In this work, we propose the use of a speech enhancement convolutional autoencoder, coupled with on-device keyword spotting, aimed at improving the trigger word detection in noisy environments. The end-to-end system learns by optimizing a linear combination of losses: a reconstruction-based loss, both at the log-mel spectrogram and at the waveform level, as well as a specific task loss that accounts for the cross-entropy error reported along the keyword spotting detection. We experiment with several neural network classifiers and report that deeply coupling the speech enhancement together with a wake-up word detector, e.g., by jointly training them, significantly improves the performance in the noisiest conditions. Additionally, we introduce a new publicly available speech database recorded for the Telefónica's voice assistant, Aura. The OK Aura Wake-up Word Dataset incorporates rich metadata, such as speaker demographics or room conditions, and comprises hard negative examples that were studiously selected to present different levels of phonetic similarity with respect to the trigger words 'OK Aura'. Keywords: speech enhancement; wake-up word; keyword spotting; deep learning; convolutional neural network
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Data augmentation for low-resource Quechua ASR improvement

  • Zevallos, Rodolfo
  • Bel Rafecas, Núria
  • Cámbara Ruiz, Guillermo
  • Farrús, Mireia
  • Luque, Jordi
Comunicació presentada a INTERSPEECH 2022, celebrat del 18 al 22 de setembre de 2022 a Inchon, Corea del Sud., Automatic Speech Recognition (ASR) is a key element in new services that helps users to interact with an automated system. Deep learning methods have made it possible to deploy systems with word error rates below 5% for ASR of English. However, the use of these methods is only available for languages with hundreds or thousands of hours of audio and their corresponding transcriptions. For the so-called low-resource languages to speed up the availability of resources that can improve the performance of their ASR systems, methods of creating new resources on the basis of existing ones are being investigated. In this paper we describe our data augmentation approach to improve the results of ASR models for low-resource and agglutinative languages. We carry out experiments developing an ASR for Quechua using the wav2letter++ model. We reduced WER by 8.73% through our approach to the base model. The resulting ASR model obtained 22.75% WER and was trained with 99 hours of original resources and 99 hours of synthetic data obtained with a combination of text augmentation and synthetic speech generation., This work has been partially supported by the Project PID2019-104512GB-I00, Ministerio de Ciencia e Innovación and Agencia Estatal de Investigación (Spain), and the INGENIOUS project from the European Union’s Horizon 2020 Research and Innovation Program under grant numbers 833435. The third author has been funded by the Agencia Estatal de Investigación (AEI), Ministerio de Ciencia, Innovación y Universidades and the Fondo Social Europeo (FSE) under grant RYC-2015-17239 (AEI/FSE, UE).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

INGENIOUS: the first responder of the future. A next generation integrated toolkit for collaborative response, increasing protection and augmenting operational capacity.

  • Institute of Communications and Computer Systems
  • Universiteit Twente
  • Deutsches Zentrum Fuer Luft – Und Raumfahrt
  • Diginext
  • Fundacion Tekniker
  • Exus
  • SINTEF
  • Center for Research & Technology Hellas
  • Totalförvarets Forskningsinstitut
  • SATWAYS
  • Alpes Lasers
  • Technische Universitaet Wien
  • Cyprus Research & Innovation Center
  • Universitat Pompeu Fabra
  • SingularLogic
  • Korea Institute of Robot & Convergence
  • Gobierno Vasco, Departamento de Seguridad
  • Hopitaux de Paris
Comunicació presentada a: SafeCorfu 2019, the 6th International Conference on Civil Protection & New Technologies, celebrat a Corfu, Grècia, del 6 al 9 de novembre de 2019., The INGENIOUS project (EU Horizon 2020) aims to develop, integrate, test, deploy, demonstrate and validate a Next Generation Integrated Toolkit (NGIT) for Collaborative Response, which ensures high level of Protection & Augmented Operational Capacity to respond to the disaster scene. This will comprise a multitude of the tools and services required: 1) for enabling protection of the first responders with respect to their health, safety and security; 2) for enhancing their operational capacities by offering them with means to conduct various response tasks and missions boosted with autonomy, automation, precise positioning, optimal utilisation of available resources and upgraded awareness and sense‐making; 3) for allowing shared response across first responders teams and disciplines by augmenting their field of view, information sharing and communications between teams and with victims. The NGIT armours the FRs at all fronts, by delivering novel, affordable, accepted and customised response tools and services as part of their uniform and as part of their operational assets. The NGIT will be provided at the service of the FRs for extensive testing and validation (at component and Toolkit levels) in the framework of a rich Training, Testing and Validation Programme – of Lab Tests (LSTs), Small‐Scale Field Tests (SSTs) and Full‐Scale Field Validations (FSXs) – towards powering the FR of the future being fully aware, fully connected and fully integrated., This work is supported by the projects INGENIOUS, which has received funding from the European Union's Horizon 2020 (H2020) programme under grant agreement No:833435.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Naturalness enhancement with linguistic information in end-to-end TTS using unsupervised parallel encoding

  • Peiró Lilja, Àlex
  • Farrús, Mireia
Comunicació presentada a Interspeech 2020 celebrat del 25 al 29 d'octubre de 2020 a Shanghai, Xina., State-of-the-art end-to-end speech synthesis models have reached levels of quality close to human capabilities. However, there is still room for improvement in terms of naturalness, related to prosody, which is essential for human-machine interaction. Therefore, part of current research has shift its focus on improving this aspect with many solutions, which mainly involve prosody adaptability or control. In this work, we explored a way to include linguistic features into the sequenceto- sequence Tacotron2 system to improve the naturalness of the generated voice. That is, making the prosody of the synthesis looking more like the real human speaker. Specifically we embedded with an additional encoder part-of-speech tags and punctuation mark locations of the input text to condition Tacotron2 generation. We propose two different architectures for this parallel encoder: one based on a stack of convolutional plus recurrent layers, and another formed by a stack of bidirectional recurrent plus linear layers. To evaluate the similarity between real read-speech and synthesis, we carried out an objective test using signal processing metrics and a perceptual test. The presented results show that we achieved an improvement in naturalness., This work is a part of the INGENIOUS project, funded by the European Union’s Horizon 2020 Research and Innovation Programme and the Korean Government under Grant Agreement No 833435. The second author has been funded by the Agencia Estatal de Investigaci´on (AEI), Ministerio de Ciencia, Innovaci ´on y Universidades and the Fondo Social Europeo (FSE) under grant RYC-2015-17239 (AEI/FSE, UE). This work has been carried out using an NVIDIA GPU Titan Xp generously provided by NVIDIA Company.