Pith. sign in

REVIEW 3 major objections 5 minor 35 references

Speaker Role and Language Diarization for Analyzing Multilingual Interviews for Language Proficiency of Older Adults

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Respondent speech ratio and intended-language usage are strong predictors of language proficiency ratings in multilingual older-adult interviews, and these signals survive when speaker-role and language labels come entirely from automatic…

desk verdict Solid applied paper with a genuinely new multilingual interview dataset and a clean pipeline; the central associations survive automatic diarization, but noisy proficiency labels and a few methodological gaps keep it from being a clean accept. read the letter →

arxiv 2608.09032 v1 pith:UF4JZX22 submitted 2026-08-10 eess.AS

Anfeng Xu , Tiantian Feng , Kevin Huang , Pranali Khobragade , Sudarsana Kadiri , Anushikha Dhankhar , Madeleine Snider , Sarah Gao
show 3 more authors
Miguel Arce Rentería Jinkook Lee Shrikanth Narayanan
This is my paper · ORCID
classification eess.AS
keywords speakerrolediarizationlanguageproficiencyassessmentWhispermultilingualinterviewsolderadultsIndiclanguagesbehavioralspeechfeatures
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that two simple, automatically measurable behaviors in multilingual interviews—how much of the conversation the respondent produces and how much of that speech is in the intended language—are reliable behavioral markers of spoken language proficiency in older adults. To test this, the authors build a two-stage Whisper-based system that first separates interviewer from respondent speech and then labels the language of each respondent segment, and they compare everything against manually annotated labels. On English, Hindi, and Marathi subsets, both behaviors show consistent positive associations with proficiency ratings, with respondent speech ratio significant in 100% of bootstrap runs and intended-language ratio in roughly 97% under fully automatic labels. The same diarization-derived features also predict Hindi proficiency scores nearly as well as Whisper speech embeddings in regression and better in binary classification, and ensembling both gives the best results. The paper's point is that these associations and predictions survive when all labels come from the automatic pipeline, making annotation-free, scalable proficiency assessment feasible.

What carries the argument

The load-bearing machinery is a two-stage diarization pipeline built on Whisper, a speech foundation model that is kept frozen and adapted with low-rank updates, followed by frame-level convolutional classifiers. Stage one classifies each 20-ms frame as interviewer, respondent, silence, or overlap; stage two classifies respondent speech frames into Hindi, English, Telugu, Marathi, Gujarati, other language, or nonverbal vocalization. The derived features—respondent speech ratio, utterance-length statistics, intended-language ratio, and vocalization ratio—are then entered into bootstrap OLS regressions and downstream ridge and logistic predictors. The comparison of oracle versus inferred labels is the experimental device that shows the pipeline preserves the signal.

What would settle it

Re-rate all 548 recordings with a small panel of expert raters and compute consensus scores; if respondent speech ratio and intended-language usage no longer show consistent positive coefficients in bootstrap regressions on those consensus scores, the central claim fails. A cheaper check would be to re-rate the Telugu subset, where current inter-rater reliability is negligible (PCC 0.079, kappa 0.049), and see whether the same associations appear.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that conversational participation and language-use patterns inferred from diarization are strong, interpretable predictors of language proficiency ratings, and that these predictors survive full automation. In bootstrap regressions on English, Hindi, and Marathi subsets, respondent speech ratio reaches statistical significance in 100% of bootstrap runs under both manual and automatic speaker-role labels, and intended-language ratio remains significant in roughly 97% of runs under fully automatic labels. For Hindi proficiency prediction, four simple diarization-derived features match or approach Whisper encoder embeddings in regression (PCC 0.441 versus 0.528) and outperform them in binary classification (61.4% versus 59.5% accuracy), with the ensemble best. The paper interprets this as evidence that respondent-centric conversational analysis, not just acoustic or lexical content, carries proficiency-relevant signal and that automatic diarization preserves it.

Load-bearing premise

The results depend on the five-point human proficiency ratings being a valid measure of true proficiency; the paper itself reports only moderate overall inter-rater reliability, and Telugu ratings are effectively unreliable, so if the ratings mostly reflect rater noise or bias, the reported associations could be artifacts.

Editorial extensions

If this is right

  • A fully automatic pipeline—speaker-role diarization followed by language diarization—can support proficiency-relevant analyses without manual labels; both statistical associations and prediction accuracy are largely retained.
  • Simple behavioral features are complementary to speech embeddings: ensembling diarization-derived features with Whisper embeddings gives the best regression and classification results, so future assessment systems can combine cheap interpretable features with learned representations.
  • Language-adapted Whisper initialization substantially reduces confusion among closely related Indic languages, with Telugu recall increasing from 28.1% to 63.5% and Marathi recall from 7.8% to 50.3%, making automatic language diarization practical for lower-resource speech conditions.
  • Nonverbal vocalization ratio takes on a stronger negative association with proficiency under automatic labels, suggesting the automatic system may capture broader disfluency patterns than manually labeled nonverbal events; this is a usable behavioral marker.
  • Because prediction performance is nearly unchanged between oracle and inferred regions, the methodology promises scalable assessment in epidemiological or clinical studies where manual annotation is expensive.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the association between speech ratio and proficiency replicates, it may partly reflect that less proficient speakers contribute less because of hesitation or avoidance; this mechanism is not tested here and could be examined by comparing first- versus second-language recordings.
  • The Telugu ratings are too unreliable to support conclusions, so the paper's language-general claim rests on Hindi, Marathi, and English; a dedicated Telugu re-rating study is a concrete next step.
  • A natural extension is to test the same diarization-derived features against criterion measures that do not depend on subjective ratings, such as clinical diagnosis or standardized bilingual proficiency batteries; if the behavioral markers predict those outcomes, the rating-noise concern would be partly bypassed.
  • Because interviewer speech occupies only about 10% of the audio here, speech ratio may be sensitive to how the interview is conducted; in other interview protocols the same feature could behave differently, so calibration across protocols is a natural next step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript introduces a Whisper-based pipeline for speaker-role and language diarization in multilingual interviewer–respondent interviews from LASI-DAD, then uses diarization-derived conversational features to analyze and predict human language-proficiency ratings. The authors curate and manually annotate 548 recordings, develop frozen-encoder Whisper models with LoRA and a CNN head, and show that VoxLect-initialized models substantially reduce language-diarization error for lower-resource Indic languages. Bootstrap OLS analyses report that respondent speech ratio and intended-language ratio are positively associated with proficiency ratings, and ridge/logistic models using diarization-derived features perform comparably to Whisper encoder embeddings on the Hindi subset. The paper's central claim is that these associations and predictions are largely preserved under fully automatic diarization, supporting scalable annotation-free proficiency assessment.

Significance. If the central claim holds, this is a solid contribution: it provides a new manually annotated multilingual older-adult interview dataset, demonstrates clear gains from language-adapted Whisper initialization for related low-resource languages, and shows that a small set of interpretable behavioral features can approximate the predictive signal of a speech foundation model. The experimental protocol has genuine strengths, including speaker-disjoint five-fold cross-validation, pooled out-of-sample predictions, oracle-versus-inferred comparisons, and bootstrap significance assessment. The dataset itself and the language-diarization improvements are valuable independent of the proficiency results. However, because the target ratings show only moderate inter-rater reliability and the headline predictors are surface behaviors that raters can easily use as fluency heuristics, the strength of the proficiency-related claims depends on additional label-validity and sensitivity analyses.

major comments (3)
  1. [§3.2.1, Table 3; §4.3] The proficiency ratings used as the target are only moderately reliable: overall quadratic-weighted kappa is 0.303, with Marathi at 0.159 and Telugu at 0.049. The exclusion of Telugu is appropriate, but Marathi remains in the Section 4.3 OLS analyses despite near-zero-to-low reliability. This matters because the two headline predictors, speech ratio and intended-language ratio, are exactly the surface behaviors a rater can use as holistic fluency heuristics; the rating descriptors in Table 2 explicitly reference slow, hesitant, and disfluent speech. The 100% bootstrap significance rates could therefore certify rater heuristics rather than a latent proficiency construct. I request sensitivity analyses using consensus or averaged ratings on the double-rated subset described in Section 3.2.1, inclusion of rater identity as a covariate, per-language association results, or an external validity criterion.
  2. [§4.3] The bootstrap OLS procedure resamples recordings, not speakers, although Section 3.1 states that a single participant may contribute multiple recordings in up to three languages. This violates the independence assumption underlying the reported significance rates and can inflate the proportion of bootstrap runs with p<0.05. A cluster bootstrap by respondent, or a mixed-effects model with a random intercept for respondent, should be used before concluding that speech ratio and intended-language ratio are strong predictors.
  3. [§6.2.2, Table 9] The claim that statistical relationships are largely preserved under automatic diarization is weakened by the nonverbal-vocalization result: under oracle annotations the coefficient is -0.144 with significance in 35.54% of runs, while under fully inferred labels it becomes -0.212 to -0.266 with significance in 94-99% of runs. The authors' explanation, that the automatic model captures a broader disfluency construct beyond true nonverbal vocalizations, may be correct, but it implies the inferred feature is not the same behavioral quantity as the annotated one. This should be validated, for example by manual inspection of segments that the model labels as nonverbal, and the 'preserved' claim should be qualified accordingly.
minor comments (5)
  1. [§4.4] The sentence 'we exclude utterance length (mean), as it has shown no significance for language proficiency scores, as shown in Section 6.3' should refer to Section 6.2.1 and Table 8, where the non-significance is reported.
  2. [Table 10] In the ensemble row, the oracle PCC and MAE columns are rendered as '0.531 0.5300.732 0.725'; the missing whitespace between the PCC and MAE values should be fixed.
  3. [§3.2.1] Calling an overall quadratic-weighted kappa of 0.303 'moderate correspondence' is optimistic; consider reporting standard reliability benchmarks and discussing the implications of the Marathi and Telugu values for the choice of target label.
  4. [Abstract and §7] The abstract and conclusion state that diarization-derived features are 'comparable to Whisper-based speech embeddings' without noting that the prediction experiments are run on the Hindi subset only; please add the qualifier.
  5. [§6.2.1] Describing speech ratio as a 'strong predictor' despite adjusted R² values around 0.15-0.17 overstates the explanatory power; consider 'robust and consistent' or report standardized effect sizes alongside the significance rates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the behavioral predictors are measured independently of proficiency labels, and proficiency predictions are evaluated on held-out folds; the only self-citation (VoxLect) is used as a pretrained encoder and is not load-bearing.

full rationale

The paper's central derivation chain is self-contained. The two headline predictors, respondent speech ratio and intended-language usage ratio, are computed from manual or automatic speaker-role and language diarization labels (Sections 3.2.2 and 4.3, Table 5), not from the proficiency ratings. The proficiency ratings come from independent human raters using a separate five-point scale (Section 3.2.1, Table 2). The bootstrap OLS analyses are associational, and the proficiency prediction experiments aggregate out-of-sample predictions across five speaker-disjoint folds (Sections 5.2 and 5.4), so no fitted parameter is renamed as a prediction. The use of the authors' own VoxLect checkpoints is a self-citation, but it functions as a pretrained encoder initialization; the claimed language-diarization improvement is demonstrated by an internal held-out comparison (Table 7) rather than imported as an external theorem, and the downstream proficiency claims do not depend on VoxLect's benchmark conclusions. The paper explicitly discloses a limitation in Section 3.2.1: inter-rater reliability is only moderate overall (PCC 0.454, quadratic-weighted kappa 0.303) and negligible for Telugu (PCC 0.079, kappa 0.049), which is why Telugu is excluded from proficiency analyses. This is a target-label validity concern, not a circularity in the derivation, because the behavioral features are not constructed from the ratings. Section 6.2.2's observation that the inferred vocalization ratio may capture broader disfluency is a confound note about feature interpretation, not a reduction of the predicted quantity to an input. No equation in the paper equates a predicted quantity to an input by construction, and no load-bearing claim rests solely on a self-citation chain.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claims rest on the integrity of the human proficiency ratings and the manual diarization labels. The paper's own data show the ratings are only moderately reliable, and the diarization annotations have no reported inter-annotator agreement. The analysis also makes post-hoc exclusion and feature-selection choices. No new entities are introduced; Whisper and VoxLect are existing models used as tools.

free parameters (3)
  • LoRA rank = 16 (Whisper-Original), 64 (VoxLect)
    Chosen by hand for training stability; affects diarization performance but not the central scientific claim.
  • Utterance merge and discard thresholds = 0.2 s
    Adjacent utterances closer than 0.2 s are merged and utterances shorter than 0.2 s are discarded when computing features from inferred regions (Section 4.4).
  • Ridge and logistic regularization strength = not reported
    The prediction models use ridge and logistic regression but the regularization hyperparameters are not specified, an unstated detail for replication.
assumptions (5)
  • domain assumption Proficiency ratings are treated as valid ground truth
    The five-point ratings are used as labels despite moderate inter-rater agreement (Section 3.2.1, Table 3).
  • domain assumption Manual speaker-role and language annotations are accurate enough to train diarization models
    No inter-annotator agreement is reported for the Praat annotation tiers (Section 3.2.2).
  • domain assumption The intended language is known and is the target language for each recording
    The intended-language ratio feature assumes the prompt language is the target; code-switching is modeled only as a deviation.
  • standard math Standard statistical and machine learning methodology is valid
    OLS, bootstrap, ridge/logistic regression, and cross-validation are accepted tools; no novel mathematical claims.
  • ad hoc to paper Data exclusion rules for low-reliability and low-count languages are appropriate
    Telugu is excluded from proficiency analyses due to negligible inter-rater reliability, and Gujarati and Others from OLS due to counts and variability (Sections 3.2.1, 4.3). These choices are post-hoc but documented.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Speaker Role and Language Diarization for Analyzing Multilingual Interviews for Language Proficiency of Older Adults." pith.science (2026). https://pith.science/paper/UF4JZX22

@misc{pith2026260809032,
  author       = {Pith},
  title        = {Pith review of: Speaker Role and Language Diarization for Analyzing Multilingual Interviews for Language Proficiency of Older Adults},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UF4JZX22}},
  note         = {Machine review of arXiv:2608.09032}
}
read the original abstract

Automatic language proficiency assessment in the context of multilingual interview-based settings remains underexplored. In this work, we develop Whisper-based speaker-role and language diarization systems to automatically extract respondent speech and characterize language usage in multilingual interviews with older adults. We further investigate whether diarization-derived conversational and language-use behaviors can support downstream language proficiency assessment. Results show that language-adapted Whisper models substantially improve language diarization performance for lower-resource and linguistically related Indian languages. Statistical analyses reveal that respondent speech ratio and intended language usage are strong predictors of proficiency ratings. Furthermore, simple diarization-derived behavioral features achieve performance comparable to Whisper-based speech embeddings for proficiency prediction, while combining both yields the best results. Importantly, both the speech and language use statistical analyses and language proficiency prediction performance remain largely preserved when using fully automatic diarization outputs, demonstrating the potential of respondent-centric conversational analysis for scalable language proficiency assessment.

Figures

Figures reproduced from arXiv: 2608.09032 by the authors.

Figure 1
Figure 1. Illustration of speaker role diarization and language diarization pipelines. 3.2.2. Speaker Role and Language Diarization Annotation We manually annotate both speaker role diarization and language diarization labels using Praat software. Each recording is annotated with three tiers: interviewer, respondent, and background-speech. Annotators create utterance￾level speech boundaries for conversational speech segments … view at source ↗
Figure 2
Figure 2. Illustration of speaker role and language diarization pipeline. The overall modeling pipeline is illustrated in [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Confusion matrices for language diarization comparing Whisper-Original and Whisper-VoxLect using Whisper￾Small. All numbers are reported in percentage (%). systems due to additional missed detection (MD) and false alarm (FA) errors introduced by the speaker-role diarization stage. Overall, Whisper-VoxLect substantially improves language diarization performance across both model sizes and VAD conditions. Under Oracle… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 27 canonical work pages

  1. [1]

    IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=

    A large-scale evaluation of speech foundation models , author=. IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=. 2024 , publisher=

    work page 2024

  2. [2]

    IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=

    End-to-end speech recognition: A survey , author=. IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=. 2023 , publisher=

    work page 2023

  3. [3]

    Computer Speech & Language , volume=

    A review of speaker diarization: Recent advances with deep learning , author=. Computer Speech & Language , volume=. 2022 , publisher=

    work page 2022

  4. [4]

    Speech Communication , volume=

    Spoken language identification: An overview of past and present research trends , author=. Speech Communication , volume=. 2025 , publisher=

    work page 2025

  5. [5]

    A Hybrid Approach to Combining Role Diarization with ASR for Professional Conversations , author=. Proc. Interspeech 2025 , pages=

    work page 2025

  6. [6]

    arXiv preprint arXiv:2601.17640 , year=

    End-to-End Joint ASR and Speaker Role Diarization with Child-Adult Interactions , author=. arXiv preprint arXiv:2601.17640 , year=

    arXiv

  7. [7]

    Interspeech , year=

    The second displace challenge: Diarization of speaker and language in conversational environments , author=. Interspeech , year=

    work page

  8. [8]

    , author=

    End to End Spoken Language Diarization with Wav2vec Embeddings. , author=. INTERSPEECH , pages=

    work page

Show all 35 references
  1. [9]

    IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=

    Implicit Self-Supervised Language Representation for Spoken Language Diarization , author=. IEEE/ACM Transactions on Audio, Speech, and Language Processing , volume=. 2024 , publisher=

    2024

  2. [10]

    arXiv preprint arXiv:2305.18881 , year=

    MERLIon CCS Challenge: A English-Mandarin code-switching child-directed speech corpus for language identification and diarization , author=. arXiv preprint arXiv:2305.18881 , year=

    arXiv

  3. [11]

    International conference on machine learning , pages=

    Robust speech recognition via large-scale weak supervision , author=. International conference on machine learning , pages=. 2023 , organization=

    2023

  4. [12]

    Journal of the American Geriatrics Society , volume=

    Design and methodology of the longitudinal aging study in India-Diagnostic assessment of dementia (LASI-DAD) , author=. Journal of the American Geriatrics Society , volume=. 2020 , publisher=

    2020

  5. [13]

    Automatic Assessment of Spoken Language Proficiency of Non-native Children , year=

    Gretter, Roberto and Matassoni, Marco and Allgaier, Katharina and Tchistiakova, Svetlana and Falavigna, Daniele , booktitle=. Automatic Assessment of Spoken Language Proficiency of Non-native Children , year=

  6. [14]

    Interspeech 2018 , pages=

    Impact of ASR performance on free speaking language assessment , author=. Interspeech 2018 , pages=. 2018 , organization=

    2018

  7. [15]

    Proficiency Assessment of L2 Spoken English Using Wav2Vec 2.0 , year=

    Bannò, Stefano and Matassoni, Marco , booktitle=. Proficiency Assessment of L2 Spoken English Using Wav2Vec 2.0 , year=

  8. [16]

    ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=

    An ASR-free fluency scoring approach with self-supervised learning , author=. ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=. 2023 , organization=

    2023

  9. [17]

    Advances in neural information processing systems , volume=

    wav2vec 2.0: A framework for self-supervised learning of speech representations , author=. Advances in neural information processing systems , volume=

  10. [18]

    ETRI Journal , volume=

    AI-based language tutoring systems with end-to-end automatic speech recognition and proficiency evaluation , author=. ETRI Journal , volume=. 2024 , publisher=

    2024

  11. [19]

    2023 , journal=

    Assessment of L2 oral proficiency using self-supervised speech representation learning , author=. 2023 , journal=

    2023

  12. [20]

    arXiv preprint arXiv:2505.21148 , year=

    Assessment of l2 oral proficiency using speech large language models , author=. arXiv preprint arXiv:2505.21148 , year=

    arXiv

  13. [21]

    Speech Communication , volume=

    Towards automatic assessment of spontaneous spoken English , author=. Speech Communication , volume=. 2018 , publisher=

    2018

  14. [22]

    arXiv preprint arXiv:2505.14648 , year=

    Vox-profile: A speech foundation model benchmark for characterizing diverse speaker and speech traits , author=. arXiv preprint arXiv:2505.14648 , year=

    arXiv

  15. [23]

    Interspeech , year=

    Exploring speech foundation models for speaker diarization in child-adult dyadic interactions , author=. Interspeech , year=

  16. [24]

    IEEE Journal of Selected Topics in Signal Processing , volume=

    Wavlm: Large-scale self-supervised pre-training for full stack speech processing , author=. IEEE Journal of Selected Topics in Signal Processing , volume=. 2022 , publisher=

    2022

  17. [25]

    Interspeech , year=

    Powerset multi-class cross entropy loss for neural speaker diarization , author=. Interspeech , year=

  18. [26]

    ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=

    Vbx for end-to-end neural and clustering-based diarization , author=. ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=. 2026 , organization=

    2026

  19. [27]

    arXiv preprint arXiv:2604.05201 , year=

    Exploring Speech Foundation Models for Speaker Diarization Across Lifespan , author=. arXiv preprint arXiv:2604.05201 , year=

    arXiv

  20. [28]

    ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=

    Leveraging self-supervised learning for speaker diarization , author=. ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , pages=. 2025 , organization=

    2025

  21. [29]

    Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V

    Voxlect: A speech foundation model benchmark for modeling dialects and regional languages around the globe , author=. Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 1 , pages=

  22. [30]

    arXiv preprint arXiv:2601.07999 , year=

    VoxCog: Towards End-to-End Multilingual Cognitive Impairment Classification through Dialectal Knowledge , author=. arXiv preprint arXiv:2601.07999 , year=

  23. [31]

    Proceedings of the 18th ACM international conference on Multimedia , pages=

    Opensmile: the munich versatile and fast open-source audio feature extractor , author=. Proceedings of the 18th ACM international conference on Multimedia , pages=

  24. [32]

    IEEE transactions on affective computing , volume=

    The Geneva minimalistic acoustic parameter set (GeMAPS) for voice research and affective computing , author=. IEEE transactions on affective computing , volume=. 2015 , publisher=

    2015

  25. [33]

    Bilingualism: Language and cognition , volume=

    What is proficiency? Characterizing spoken language proficiency in older Spanish-English bilinguals , author=. Bilingualism: Language and cognition , volume=. 2026 , publisher=

    2026

  26. [34]

    IEEE transactions on audio, speech, and language processing , volume=

    Spoken language derived measures for detecting mild cognitive impairment , author=. IEEE transactions on audio, speech, and language processing , volume=. 2011 , publisher=

    2011

  27. [35]

    Journal of the American Geriatrics Society , volume=

    Introduction to LASI-DAD: the longitudinal aging study in India-diagnostic assessment of dementia , author=. Journal of the American Geriatrics Society , volume=

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.