Nonlinear waveform distortion: Assessment and detection of clipping on speech data and systems

Hansen, John H. L.; Stauffer, Allen; Xia, Wei

doi:10.1016/j.specom.2021.07.007

Cited by 15 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further semi-automated quality analysis after noise suppression was performed to ensure high-quality data were analyzed. Thresholds were signal to noise ratio (SNR) > 30Db [ 7 ], clipping in fewer than 1% of data samples [ 15 ], and no unusual patterns of noise as evident by visual inspection of spectrograms (e.g., narrowband noise). These steps were performed by trained and experienced research assistants.…”

Section: Methodsmentioning

confidence: 99%

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Simmatis,

Robin,

Spilka

et al. 2024

BioMed Eng OnLine

View full text Add to dashboard Cite

Automatic speech assessments have the potential to dramatically improve ALS clinical practice and facilitate patient stratification for ALS clinical trials. Acoustic speech analysis has demonstrated the ability to capture a variety of relevant speech motor impairments, but implementation has been hindered by both the nature of lab-based assessments (requiring travel and time for patients) and also by the opacity of some acoustic feature analysis methods. These challenges and others have obscured the ability to distinguish different ALS disease stages/severities. Validation of automated acoustic analysis tools could enable detection of early signs of ALS, and these tools could be deployed to screen and monitor patients without requiring clinic visits. Here, we sought to determine whether acoustic features gathered using an automated assessment app could detect ALS as well as different levels of speech impairment severity resulting from ALS. Speech samples (readings of a standardized, 99-word passage) from 119 ALS patients with varying degrees of disease severity as well as 22 neurologically healthy participants were analyzed, and 53 acoustic features were extracted. Patients were stratified into early and late stages of disease (ALS-early/ALS-E and ALS-late/ALS-L) based on the ALS Functional Ratings Scale-Revised bulbar score (FRS-bulb) (median [interquartile range] of FRS-bulbar scores: 11[3]). The data were analyzed using a sparse Bayesian logistic regression classifier. It was determined that the current relatively small set of acoustic features could distinguish between ALS and controls well (area under receiver-operating characteristic curve/AUROC = 0.85), that the ALS-E patients could be separated well from control participants (AUROC = 0.78), and that ALS-E and ALS-L patients could be reasonably separated (AUROC = 0.70). These results highlight the potential for automated acoustic analyses to detect and stratify ALS.

show abstract

Section: Methodsmentioning

confidence: 99%

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Simmatis,

Robin,

Spilka

et al. 2024

BioMed Eng OnLine

View full text Add to dashboard Cite

show abstract

“…Due to the caregiver being within variable distance of their microphones, some clipping was identified in a sample of the microphone recordings. A stringent clipping identification algorithm was used (see SM 1.1 and Fig.S5) to remove parts of the microphone data where clipping occurred 92 . Vocalisations where any clipping was identified were set to missing data points.…”

Section: Methodsmentioning

confidence: 99%

“…The clipping algorithm was based on that outlined by 1 . First, points in the speech signal reaching the maximum or minimum amplitude were identified.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Endogenous oscillatory rhythms and interactive contingencies jointly influence infant attention during early infant-caregiver interaction

Phillips,

Goupil,

Whitehorn

et al. 2023

Preprint

View full text Add to dashboard Cite

Almost all early cognitive development takes place in social contexts. At the moment, however, we understand little about the neural and cognitive mechanisms that drive infant attention during social interactions. Recording EEG during naturalistic caregiver-infant interactions (N=66), we compare two different accounts. Attentional scaffolding perspectives emphasise the role of the caregiver in structuring the interaction, whilst active learning models focus on motivational factors, endogenous to the infant, that guide their attention. Our results show that, already by 12 months, intrinsic cognitive processes control infant attention: fluctuations in endogenous oscillatory neural activity associated with changes in infant attentiveness, and predicted the length of infant attention episodes towards objects. In comparison, infant attention was not forwards-predicted by caregiver gaze, or modulations in the spectral and temporal properties of the caregiver's speech. Instead, caregivers rapidly modulated their behaviours in response to changes in infant attention and cognitive engagement, and greater reactive changes associated with longer infant attention. Our findings suggest that shared attention develops through interactive by asymmetric, infant-led processes that operate across the caregiver-child dyad.

show abstract

“…Further semi-automated quality analysis after noise suppression was performed to ensure highquality data were analyzed. Thresholds were signal to noise ratio (SNR) > 30Db (Deliyski et al, 2005), clipping in fewer than 1% of data samples (Hansen et al, 2021), and no unusual patterns of noise as evident by visual inspection of spectrograms (e.g., narrowband noise). These steps were performed by trained and experienced research assistants.…”

Section: Methodsmentioning

confidence: 99%

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Simmatis,

Robin,

Spilka

et al. 2023

Preprint

View full text Add to dashboard Cite

Home-based speech assessments have the potential to dramatically improve ALS clinical practice and facilitate patient stratification for ALS clinical trials. Acoustic speech analysis has demonstrated the ability to capture a variety of relevant speech motor impairments, but implementation has been hindered by both the nature of lab-based assessments (requiring travel and time for patients) and also by the opacity of some acoustic feature analysis methods. Furthermore, these challenges and others have obscured the ability to distinguish different ALS disease stages/severities. Validation of remote-capable acoustic analysis tools could enable detection of early signs of ALS, and these tools could be deployed to screen and monitor patients without requiring clinic visits. Here, we sought to determine whether acoustic features gathered using a remote-capable assessment app could detect ALS as well as different levels of speech impairment severity resulting from ALS. Speech samples (readings of a standardized, 99-word passage) from 119 ALS patients with varying degrees of disease severity as well as 22 neurologically healthy participants were analyzed, and 53 acoustic features were extracted. Patients were stratified into early and late stages of disease (ALS-early/ALS-E and ALS-late/ALS-L) based on the ALS Functional Ratings Scale - Revised bulbar score (FRS-bulb). Data were analyzed using a sparse Bayesian logistic regression classifier. It was determined that the current relatively small set of acoustic features could distinguish between ALS and controls well (area under receiver operating characteristic curve/AUROC = 0.85), that the ALS-E patients could be separated well from control participants (AUROC = 0.78), and that ALS-E and ALS-L patients could be reasonably separated (AUROC = 0.70). These results highlight the potential for remote acoustic analyses to detect and stratify ALS.

show abstract

Nonlinear waveform distortion: Assessment and detection of clipping on speech data and systems

Cited by 15 publications

References 11 publications

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Endogenous oscillatory rhythms and interactive contingencies jointly influence infant attention during early infant-caregiver interaction

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis

Contact Info

Product

Resources

About