Audio ML researcher on efficient pitch estimation and audio restoration for edge devices; early-stage profile
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-09) 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: the snapshot reports h-index 0 and i10-index 0, so publication volume cannot be meaningfully assessed. This axis would measure total output volume, including the count of works and sustained productivity over time.
Impact—
Not yet scored: citation-based measures (h-index, i10-index, total citations) are absent or zero in the current OpenAlex snapshot. This axis would measure citation influence — how often the researcher's work is taken up by peers.
Focus0/100
The radar focus score is 0, reflecting output split evenly across two adjacent fields: audio signal processing / music information retrieval (0.5) and audio signal processing / machine learning (0.5). Underlying OpenAlex topics nevertheless cluster tightly on audio: Speech and Audio Processing, Music and Audio Processing, and Music Technology and Sound Studies each carry 4 works, with Speech Recognition and Synthesis and Neural Networks and Applications at 1 each.
Collaboration22/100
Collaboration scores 22, based on just 2 distinct coauthors in the recent window. Venkat Suprabath Bitra and Yi-Hung Kan are the recurring collaborators recorded. On both reviewed 2026 arXiv papers Homayoon Beigi appears in the last author position (not corresponding), a placement often associated with senior or supervisory roles, though no affiliation data confirms this. The small collaborator count indicates a compact, largely stable working group rather than a broad network.
Activity52/100
Activity scores 52. The OpenAlex snapshot records 5 works in the last three years, with countsByYear showing all 5 appearing in 2026 — a single-year burst rather than a multi-year trend. Both cataloged papers are 2026 arXiv preprints, so current output is recent and concentrated; a longer-run publishing cadence cannot be established from the available window.
Reproducibility55/100
Reproducibility scores 55 (reproMean 55 across 2 scored papers). Both reviewed papers are rated medium on repro-readiness. The pitch-estimation paper is described as rigorous but without external baselines or released code; the vinyl-defect paper offers unusually honest ablations yet evaluates on only three held-out recordings. The pattern is mixed but transparent, with clear gaps in external validation and public code.
Research summary
Homayoon Beigi is an audio signal processing researcher whose recent work sits at the intersection of music information retrieval and applied machine learning: the OpenAlex field distribution splits evenly, 50/50, between audio signal processing / music information retrieval and audio signal processing / machine learning. The snapshot records 5 works in the last three years, all dated 2026, with topic counts concentrated on Speech and Audio Processing (4), Music and Audio Processing (4), and Music Technology and Sound Studies (4), alongside single works in Speech Recognition and Synthesis and Neural Networks and Applications. Two reviewed 2026 arXiv preprints define the cataloged record, both carrying the researcher in the last author position and neither designated corresponding. The first applies rank-9 low-rank frequency convolution and noise-range augmentation to real-time pitch estimation on edge devices, cutting model size by 35.9% at no accuracy cost. The second builds a two-stage expert-plus-Viterbi detector for vinyl defect regimes, reaching 0.730 five-class macro F1 on synthetic data. Evolve rates both papers medium on repro-readiness, scoring reproducibility at 55 overall. Collaboration is compact — 2 distinct coauthors, with Venkat Suprabath Bitra and Yi-Hung Kan recurring — and activity scores 52. Citation-based output and impact remain unscored.
Collaboration pattern
The snapshot lists just 2 distinct coauthors, and collaboration scores 22 — a compact working pattern. Venkat Suprabath Bitra and Yi-Hung Kan are the recurring collaborators. On both reviewed 2026 arXiv papers the researcher sits in the last author position (not designated corresponding), which often signals a supervisory or senior slot, though no affiliation data confirms the role. No cross-institution or large-team collaborations are visible in this window.
Research keywords
Highlights
35.9% model compression for real-time edge pitch estimation
Expert-plus-Viterbi detector for vinyl defect regimes
Timeline
- 2026Released the arXiv preprint 'Low-Rank Frequency Convolution and Noise-Range Augmentation for Real-Time Pitch Estimation on Edge Devices', as last author.
- 2026Released the arXiv preprint 'Signal-Informed Temporal Routing for Vinyl Defect Regime Detection', as last author.
- 2026OpenAlex countsByYear records 5 works in 2026, the only year with recorded output in this snapshot.
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.
- Bibliometric figures come from an OpenAlex snapshot only; no ORCID, institutional affiliation, or country was available, so this profile is name-resolved from public bibliographic data.
- Output and impact dimensions are unscored: h-index and i10-index are 0 in the snapshot and no citation counts were provided.
- countsByYear shows works only in 2026 and only 2 papers carry Evolve reviews; career length, publishing trends, and unreviewed work are not captured.
- The frequent-coauthor list and topic counts reflect only the works indexed in this recent window.