Intensity feature for speech stress detection

Czap, László; Pintér, Judit Mária

doi:10.1109/carpathiancc.2015.7145052

Cited by 7 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These data can have the form of categorical (e.g., speakers style or intention) or numerical information (e.g., speech pauses count and lengths) or can be represented by a time series (e.g., speech rate, voice vibration or modulation). Moreover, we can obtain additional information from the statistical analysis of such speech parameters through the lenses of affective computing, such as stress detection [7] or reasoning about the speaker's concealed emotions [43]. Some feedback will be provided in real-time, and the rest will be shown retrospectively as a report after the presentation.…”

Section: Results Reportingmentioning

confidence: 99%

Towards Multimodal VR Trainer of Voice Emission and Public Speaking -Work-in-Progress

loras-Cvbulska¹,

Hekiert

Cybulski³

et al. 2023

2023 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

View full text Add to dashboard Cite

Section: Results Reportingmentioning

confidence: 99%

Towards Multimodal VR Trainer of Voice Emission and Public Speaking -Work-in-Progress

loras-Cvbulska¹,

Hekiert

Cybulski³

et al. 2023

2023 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

View full text Add to dashboard Cite

“…The F0 contour reflects the amplitude of a change in log energy in the voiced area of a syllable. Boundary identification is a method for calculating stress [40].…”

Section: Prosodic Featuresmentioning

confidence: 99%

Prosodic Feature-Based Discriminatively Trained Low Resource Speech Recognition System

et al. 2022

View full text Add to dashboard Cite

Speech recognition has been an active field of research in the last few decades since it facilitates better human–computer interaction. Native language automatic speech recognition (ASR) systems are still underdeveloped. Punjabi ASR systems are in their infancy stage because most research has been conducted only on adult speech systems; however, less work has been performed on Punjabi children’s ASR systems. This research aimed to build a prosodic feature-based automatic children speech recognition system using discriminative modeling techniques. The corpus of Punjabi children’s speech has various runtime challenges, such as acoustic variations with varying speakers’ ages. Efforts were made to implement out-domain data augmentation to overcome such issues using Tacotron-based text to a speech synthesizer. The prosodic features were extracted from Punjabi children’s speech corpus, then particular prosodic features were coupled with Mel Frequency Cepstral Coefficient (MFCC) features before being submitted to an ASR framework. The system modeling process investigated various approaches, which included Maximum Mutual Information (MMI), Boosted Maximum Mutual Information (bMMI), and feature-based Maximum Mutual Information (fMMI). The out-domain data augmentation was performed to enhance the corpus. After that, prosodic features were also extracted from the extended corpus, and experiments were conducted on both individual and integrated prosodic-based acoustic features. It was observed that the fMMI technique exhibited 20% to 25% relative improvement in word error rate compared with MMI and bMMI techniques. Further, it was enhanced using an augmented dataset and hybrid front-end features (MFCC + POV + Fo + Voice quality) with a relative improvement of 13% compared with the earlier baseline system.

show abstract

“…A Zigbee, Z-wave átjárókkal képes kommunikálni és ennek hatására vezérelni a hozzá csatlakoztatott eszközöket. A mi házunkban az Amazon Alexa szolgáltatása egy Amazon Echo Dot 3 eszköz révén lesz jelen [1]. Ezt az eszközt a 3. ábrámon láthatjuk.…”

Section: Amazon Alexaunclassified

Beszédalapú interfész integrálása okosotthon vezérlésébe

Varga

2019

MDT

View full text Add to dashboard Cite

Az okosotthonok dinamikus fejlődésével egyre nagyobb teret nyernek az okos megoldások a háztartásban. Az informatika fejlődésével a hangasszisztensek is egyre elterjedtebbé váltak. Több nagy cég is rendelkezik hangasszisztenssel amelyeket integrálni lehet egy 21. századi okosotthonba. Egy hangasszisztens integrálásával sokkal kényelmesebben és természetesebben lehet az okosotthont vezérelni, mint bármely más eszköz segítségével, hiszen a beszéd az emberek közötti legalapvetőbb kommunikációs forma. Az egyetemünk részt vett a Solar Decathlon versenyen, amelyen a házat többek kötött hangasszisztenssel is lehet vezérelni. A Solar Decathlon Europe 2019 nemzetközi egyetemi innovációs házépítő verseny, amelyen az egyetemünk a Pécsi tudományegyetemmel és az algériai Blidai egyetemmel konzorciumban vett részt. A célom egy olyan rendszer megalkotása volt, amely képes a hangasszisztenst a ház vezérlésébe integrálni.

show abstract

Intensity feature for speech stress detection

Cited by 7 publications

References 2 publications

Towards Multimodal VR Trainer of Voice Emission and Public Speaking -Work-in-Progress

Towards Multimodal VR Trainer of Voice Emission and Public Speaking -Work-in-Progress

Prosodic Feature-Based Discriminatively Trained Low Resource Speech Recognition System

Beszédalapú interfész integrálása okosotthon vezérlésébe

Contact Info

Product

Resources

About