Impact of ASR on Alzheimer’s Disease Detection: All Errors are Equal, but Deletions are More Equal than Others

Balagopalan, Aparna; Shkaruta, Ksenia; Novikova, Jekaterina

doi:10.18653/v1/2020.wnut-1.21

Cited by 4 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cepstral Mel-Frequency Components (MFCC), among others [10]. Different combinations of these features were used in multiple previous studies when detecting AD from the speech collected via picture description tasks [4,5,11,12]. These works have provided clear evidence on the potential of using simple spoken tasks and conventional acoustic features to automatically assess early dementia and its progression as well as to demonstrate that technology allows automatic detection of AD.…”

Section: Extracting Conventional Acoustic Features From Speechmentioning

confidence: 99%

See 1 more Smart Citation

Comparing Acoustic-based Approaches for Alzheimer's Disease Detection

Balagopalan¹,

Novikova²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Robust strategies for Alzheimer's disease (AD) detection is important, given the high prevalence of AD. In this paper, we study the performance and generalizability of three approaches for AD detection from speech on the recent ADReSSo challenge dataset:1) using conventional acoustic features 2) using novel pre-trained acoustic embeddings 3) combining acoustic features and embeddings. We find that while feature-based approaches have a higher precision, classification approaches relying on the combination of embeddings and features prove to have a higher, and more balanced performance across multiple metrics of performance. Our best model, using such a combined approach, outperforms the acoustic baseline in the challenge by 2.8%.

show abstract

Section: Extracting Conventional Acoustic Features From Speechmentioning

confidence: 99%

“…This is not surprising, as conventional acoustic features were engineered based on substantial domain knowledge in the area of AD detection. Multiple previous studies [4,5,11,12] were using these features as the ones that characterize the speech of patients with AD the best.…”

Section: Generalizabilitymentioning

confidence: 99%

Comparing Acoustic-based Approaches for Alzheimer's Disease Detection

Balagopalan¹,

Novikova²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Linguistically-based features can be very informative to detect AD [16,17,18]; however, they rely on the performance of an ASR system, which are prone to introduce errors, or human transcriptions, which are very time-consuming, and they are language dependant. The more elaborate the language analysis is, more sensible to ASR errors will be, making a real system less robust.…”

Section: Alzheimer Disease Detection From Speechmentioning

confidence: 99%

Influence of ASR and Language Model on Alzheimer's Disease Detection

Codina-Filbà¹,

Cámbara²,

Luque³

et al. 2021

Preprint

View full text Add to dashboard Cite

Alzheimer's Disease is the most common form of dementia. Automatic detection from speech could help to identify symptoms at early stages, so that preventive actions can be carried out. This research is a contribution to the ADReSSo Challenge, we analyze the usage of a SotA ASR system to transcribe participant's spoken descriptions from a picture. We analyse the loss of performance regarding the use of human transcriptions (measured using transcriptions from the 2020 ADReSS Challenge). Furthermore, we study the influence of a language modelwhich tends to correct non-standard sequences of words-with the lack of language model to decode the hypothesis from the ASR. This aims at studying the language bias and get more meaningful transcriptions based only on the acoustic information from patients. The proposed system combines acoustic -based on prosody and voice quality-and lexical features based on the first occurrence of the most common words. The reported results show the effect of using automatic transcripts with or without language model. The best fully automatic system achieves up to 76.06 % of accuracy (without language model), significantly higher, 3 % above, than a system employing word transcriptions decoded using general purpose language models.

show abstract

“…There can be three types of language perturbations at the word level: insertions, deletions, and substitutions on words. (Balagopalan et al, 2019) showed that deletions are more affected (significantly) than insertions and substitutions, so we likewise focus on deletions. Following Balagopalan et al ( 2019), we artificially add deletion errors to original individual text samples at predefined levels of 20%, 40%, 60%, and 80%.…”

Section: Altering Text Samplesmentioning

confidence: 99%

Lexical Features Are More Vulnerable, Syntactic Features Have More Predictive Power

Novikova

Balagopalan

Shkaruta

et al. 2019

Proceedings of the 5th Workshop on Noisy User-Generated Text (W-Nut 2019)

Self Cite

View full text Add to dashboard Cite

Understanding the vulnerability of linguistic features extracted from noisy text is important for both developing better health text classification models and for interpreting vulnerabilities of natural language models. In this paper, we investigate how generic language characteristics, such as syntax or the lexicon, are impacted by artificial text alterations. The vulnerability of features is analysed from two perspectives: (1) the level of feature value change, and (2) the level of change of feature predictive power as a result of text modifications. We show that lexical features are more sensitive to text modifications than syntactic ones. However, we also demonstrate that these smaller changes of syntactic features have a stronger influence on classification performance downstream, compared to the impact of changes to lexical features. Results are validated across three datasets representing different text-classification tasks, with different levels of lexical and syntactic complexity of both conversational and written language.

show abstract

Impact of ASR on Alzheimer’s Disease Detection: All Errors are Equal, but Deletions are More Equal than Others

Cited by 4 publications

References 23 publications

Comparing Acoustic-based Approaches for Alzheimer's Disease Detection

Comparing Acoustic-based Approaches for Alzheimer's Disease Detection

Influence of ASR and Language Model on Alzheimer's Disease Detection

Lexical Features Are More Vulnerable, Syntactic Features Have More Predictive Power

Contact Info

Product

Resources

About