University of Surrey researcher applying explainable AI and hierarchical clustering to speaker recognition embeddings
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
No publication-volume metric is available yet. This axis would measure total research output over time — number of papers per year, publication venues, and author roles — so productivity cannot currently be scored or compared against peers.
Impact—
No citation-based data is available yet. This axis would measure scholarly influence — citations to the reviewed papers, downstream uptake and referencing context — so no impact score can be reported for this profile.
Focus0/100
The platform scores focus at 0/100. The underlying field distribution splits evenly, 50% 'speech processing / speaker recognition (explainable AI)' and 50% 'speech technology and machine learning (explainable AI)', across 2 catalogued papers. In practice the two labels describe adjacent topics — both papers target explainability of speaker embeddings — so the low score reflects an even two-way split on a very small sample rather than demonstrated breadth across unrelated fields.
Collaboration22/100
Collaboration scores 22/100, based on only 2 distinct coauthors in the snapshot. The recurring partners are Yanze Xu and Mark D. Plumbley, who co-appear on the 2026 explainable-speaker-embedding papers. On both catalogued outputs the researcher sits in a middle author position and is not the corresponding author, so the record shows a compact, stable co-author cluster within the University of Surrey group rather than a broad external network.
Activity—
No temporal publication data is available yet. This axis would measure research cadence — papers per year, gaps and recency of output — so the current activity level cannot be scored from the snapshot.
Reproducibility38/100
Reproducibility readiness averages 38/100 across the 2 reviewed papers. 'Interpreting hierarchical organisation of speaker embeddings' is rated medium, with clearly defined HCCM and L-score procedures, though it uses a single model and dataset with no clustering baselines; 'Exploring Second-Order Pattern Recognition in Speaker Recognition' is rated low, with a heuristic Gradient Cost measure and thin evaluation. Releasing code, adding baselines and multi-dataset evaluation would raise this axis.
Research summary
Wenwu Wang is a researcher at the University of Surrey whose recent work sits at the intersection of speech technology and explainable artificial intelligence. The Evolve catalog contains two of the researcher's 2026 arXiv papers, both addressing how structure inside speaker embeddings can be made interpretable. The first introduces HCCM and an L-score for interpreting SLINK single-linkage hierarchies of speaker embeddings; the second extends the same SLINK-plus-HCCM framework with a new HCNA method to discover and assign latent second-order patterns, adding a heuristic Gradient Cost measure. Across the catalogued record the field distribution splits evenly (50%/50%) between 'speech processing / speaker recognition (explainable AI)' and 'speech technology and machine learning (explainable AI)', and the researcher appears in a middle author position on both papers without corresponding-author role. The observed collaboration footprint is compact: 2 distinct coauthors, with Yanze Xu and Mark D. Plumbley the most frequent. Evolve rates reproducibility readiness at a mean of 38/100 across the two papers (one medium, one low), reflecting single-model/single-dataset evaluation and thin baselines. Output, impact and activity metrics are not yet available in the platform's snapshot.
Collaboration pattern
The visible collaboration footprint is small and stable: 2 distinct coauthors, with Yanze Xu and Mark D. Plumbley appearing most frequently alongside the researcher on the 2026 explainable-speaker-embedding papers. On both catalogued outputs the researcher holds a middle author position without corresponding-author status, consistent with contributing within an established University of Surrey team rather than leading an independent group. No external-network breadth can be assessed from this snapshot.
Research keywords
Highlights
HCCM and an L-score for reading SLINK hierarchies of speaker embeddings
Discovering second-order patterns in speaker recognition with HCNA
Timeline
- 2026Publishes 'Interpreting hierarchical organisation of speaker embeddings' on arXiv, introducing HCCM and an L-score for interpreting SLINK hierarchies of speaker embeddings; middle author, non-corresponding.
- 2026Publishes 'Exploring Second-Order Pattern Recognition in Speaker Recognition' on arXiv, adding the HCNA method and a Gradient Cost measure on top of SLINK+HCCM; middle author, non-corresponding.
- 2026Both catalogued outputs fall in explainable-AI-oriented speech research, with 2 distinct coauthors — Yanze Xu and Mark D. Plumbley — and a mean reproducibility readiness of 38/100 across the two papers.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Radar scores are computed from a small platform bibliometric snapshot (2 catalogued papers, 2 distinct coauthors), so values are indicative rather than definitive.
- Output, impact and activity dimensions lack sufficient underlying data and are reported as missing, not estimated.
- 'Wenwu Wang' is a common name; this profile is name-resolved from public bibliographic data and the stated affiliation, and no ORCID is provided for identity confirmation.
- Both reviewed items are 2026 arXiv preprints with middle, non-corresponding author positions; peer-reviewed venues and independent citation verification are outside this snapshot.