Veteran music & audio computing researcher at Universitat Pompeu Fabra, now shaping evaluation of generative music AI
3
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
46
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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 output score is available yet — the radar lacks the data behind this axis. It would measure annual and cumulative publication volume, venue spread, and author-position share; only raw countsByYear figures are on file.
Impact—
No impact score is available yet — citation-linked fields are incomplete in the snapshot. This axis would measure citation counts, h-index-based influence and uptake of his datasets and tools; h-index 46 and i10-index 156 are recorded but unscored.
Focus100/100
Focus scores 100. Topic counts are tightly concentrated in music and audio: Music and Audio Processing (413 records) and Music Technology and Sound Studies (350) dominate, followed by Musicology and Musical Analysis (156), Speech and Audio Processing (91) and Speech Recognition and Synthesis (70). The profile shows a coherent, stable research identity with minimal drift.
Collaboration51/100
Collaboration scores 51 — mid-range — with 12 distinct recent co-authors. The network combines long-running collaborators (Emilia Gómez, Joan Serrà, Frederic Font, Martín Rocamora) with a second recurring cluster (Yuki Mitsufuji, Woosung Choi, Wei-Hsiang Liao) and others such as Roser Batlle-Roca, Laura Ibáñez-Martínez, Esteban Gutiérrez, Lonce Wyse, Fabio Morreale. Across the 2026 catalog papers he appears in last or middle author positions, consistent with senior supervisory co-authorship.
Activity100/100
Activity scores 100. The countsByYear series runs unbroken from 1986 to 2026: 21 works in 2023 and 2024, 26 in 2025 and 20 already in 2026, summing to 67 works over the last three years. After a four-decade arc that peaked at 38 works in 2018, output remains consistently high — hence the top-of-scale score.
Reproducibility100/100
Reproducibility scores 100 (reproMean 100), but on a single evaluated paper (reproPapers 1): SCAPES, whose 36M-parameter flow prior over EnCodec latents earned a high repro-readiness rating. The other two 2026 catalog papers are rated medium, so the perfect score should be read as a promising early signal, not a settled record.
Research summary
Xavier Serra, at Universitat Pompeu Fabra in Spain, is a long-standing researcher in music and audio computing whose OpenAlex-recorded output spans four decades, from a first recorded publication in 1986 to a steady 20-26 works per year across 2023-2026. His profile carries an h-index of 46 and an i10-index of 156, with topics concentrated in Music and Audio Processing (413 records), Music Technology and Sound Studies (350), Musicology and Musical Analysis (156), Speech and Audio Processing (91) and Speech Recognition and Synthesis (70). Recent activity remains high — 67 works in the last three years — earning perfect activity and focus scores on Evolve's radar. The three 2026 papers in Evolve's catalog show a current emphasis on evaluating generative music and audio AI: SCAPES, a 36M-parameter flow prior over EnCodec latents for environmental sound; MATCHA, 300 attribute-level music-similarity triplets judged by 83 experts; and MusGU+, a musician-centered framework scoring 10 music-AI systems on 15 criteria. He appears as a senior co-author (last on SCAPES, middle elsewhere) within a compact network of 12 recent collaborators including Emilia Gómez, Joan Serrà, Frederic Font and Yuki Mitsufuji. Two gaps remain: output and impact scores are not yet computed, and the perfect reproducibility score rests on a single evaluated paper.
Collaboration pattern
His recent network is compact — 12 distinct co-authors in the current window — but the frequent-coauthor list spans a long arc, from longtime colleagues such as Emilia Gómez, Joan Serrà and Frederic Font to other collaborators including Roser Batlle-Roca, Laura Ibáñez-Martínez and Esteban Gutiérrez, alongside a recurring cluster with Yuki Mitsufuji, Woosung Choi and Wei-Hsiang Liao. On the 2026 catalog papers he appears as last author once and middle author twice, consistent with senior co-authorship.
Research keywords
Highlights
SCAPES: a lightweight flow prior for environmental sound
MATCHA: 300 expert-judged music-similarity triplets
MusGU+: a musician-centered yardstick for generative music AI
Timeline
- 1986Earliest recorded publication in the OpenAlex countsByYear series (1 work), marking the start of a four-decade output record.
- 2014Annual output climbs to 37 works, part of a steep 2011-2018 rise from 14 to 38 works per year.
- 2018Peak annual count in the series: 38 works.
- 2026Three arXiv preprints in Evolve's catalog — SCAPES, MATCHA and MusGU+ — with 20 works recorded so far this year.
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 (h-index 46, i10-index 156, countsByYear, topics) come from an OpenAlex-sourced snapshot only; no other citation databases were consulted.
- Output and impact radar scores are missing in the source data; this profile reports no substitute or estimated scores for them.
- The reproducibility score of 100 derives from a single evaluated paper (reproPapers 1) and may not generalize.
- Identity is name-resolved from public bibliographic data (OpenAlex, ORCID 0000-0003-1395-2345); only the three cataloged 2026 papers received full Evolve review.