Early-career researcher in full-duplex spoken dialogue: turn-taking benchmarks and synthesized dialogue data
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) 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—
Not yet scored: this identity's indexed output is too small to measure reliably. This axis would track publication volume, venue quality, and authorship position as the record accumulates beyond its first 2026 entries.
Impact—
Not yet scored: no citation evidence is available yet for these 2026 preprints. This axis would measure citations, field-normalized influence, and uptake of the released benchmarks and datasets by other research groups.
Focus0/100
The focus score of 0 reflects an even 0.5/0.5 split between two field labels that both denote speech processing / spoken dialogue systems — label fragmentation rather than genuine breadth. Topic counts confirm a narrow agenda: speech recognition and synthesis (4), speech and dialogue systems (4), speech and audio processing (3), with a single work on discourse and communication strategies, all centered on full-duplex turn-taking.
Collaboration60/100
Collaboration scores 60, based on 19 distinct coauthors across a four-work record — a wide network for such an early stage. Recurring partners include Qingxiang Guo, Yingxin Shang, Hongwei Cai, Weixu Wang, and Shuofeng Zhao, suggesting a stable team. In both reviewed papers the researcher sits in a middle author position, consistent with contributing to shared infrastructure such as data and benchmarks rather than solo leadership.
Activity47/100
Activity scores 47. OpenAlex records 4 works, all in 2026, matching a recent-3-year count of 4 — a burst concentrated in a single year with no earlier indexed history. All catalogued outputs are arXiv preprints rather than peer-reviewed publications, which tempers the score; the first-year pace is steady but the visible track record is short.
Reproducibility38/100
Reproducibility scores 38 (reproMean 38 across 2 reviewed papers). ECHO rates medium: its matched-insertion design supports controlled comparison, but Evolve notes label-dependent interruption acoustic processing and the absence of a gain-free or waveform-reuse control. DuplexDrama rates low: the 800-hour synthesis pipeline is documented, yet without human perceptual validation the quality of the generated dialogue is hard to verify independently.
Research summary
Wenke Fan is an early-career researcher working on full-duplex spoken dialogue systems; this profile is assembled from OpenAlex bibliographic records and Evolve's independent reviews of two 2026 arXiv preprints. OpenAlex attributes 4 works to this identity, all dated 2026, so the record is very new: h-index and i10-index are recorded as zero, and the output and impact axes are left unscored rather than estimated. The topical distribution is tightly clustered — speech recognition and synthesis (4), speech and dialogue systems (4), speech and audio processing (3), plus one work on language, discourse and communication strategies. Both reviewed papers target evaluation and data infrastructure for full-duplex dialogue. ECHO introduces a matched-insertion paired benchmark that exposes a YIELD bias in turn-taking models, though Evolve flags label-dependent interruption acoustic processing and a missing gain-free or waveform-reuse control. DuplexDrama claims the first 800-hour TTS-synthesized full-duplex dialogue dataset spanning personas, behaviors, expressive speech, and sound events, with a solid pipeline but no human perceptual validation. The researcher holds a middle author position on both.
Collaboration pattern
The record lists 19 distinct coauthors across 4 works — a broad network for a first-year researcher. Qingxiang Guo, Yingxin Shang, Hongwei Cai, Weixu Wang, Shuofeng Zhao, Dawei Yang, Zhiyang Zhou, Lin Yang, Shuran Zhou, and Yang Song recur, implying a stable group model. In both reviewed papers the researcher holds a middle author position without corresponding-author status, consistent with building shared data and benchmark infrastructure within a team.
Research keywords
Highlights
ECHO: matched-contrast benchmark exposes YIELD bias in full-duplex turn-taking
DuplexDrama: claims the first 800-hour TTS full-duplex dialogue dataset
Timeline
- 2026OpenAlex record opens: 4 works indexed for this identity, all dated to this year.
- 2026DuplexDrama preprint posted on arXiv — a synthesized 800-hour full-duplex dialogue dataset with personas, behaviors, expressive speech, and sound events.
- 2026ECHO preprint posted on arXiv — a matched-contrast benchmark for context-sensitive turn-taking in full-duplex dialogue.
- 2026Both preprints are reviewed by Evolve, forming the evidence base of this early-stage profile.
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.
- Bibliometrics are derived from OpenAlex only; identity resolved via ORCID 0009-0005-8808-648X. Institution and country are not available in the source data.
- h-index and i10-index are recorded as 0; the output and impact dimensions are left unscored because the 2026-dated record is too new for citation evidence.
- Only 2 of the 4 indexed works have Evolve reviews; both are arXiv preprints, so peer-review status is not established.
- "Wenke Fan" is a common Chinese name; this profile is name-resolved from public bibliographic data and may exclude same-name work indexed elsewhere.