Speech and NLP researcher at Brno University of Technology, active across ASR evaluation and speech emotion recognition
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
27
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 snapshot lacks the full publication-type and venue-level breakdown needed to compute an output score; the axis would measure the volume and consistency of research production. Raw signals on file, such as h-index 27 and i10-index 71, have not been converted into a score.
Impact—
Not yet scored. Impact would measure citation uptake and influence relative to field norms; the snapshot provides only raw h-index 27 and i10-index 71, without citation distributions or field-normalized rates, so no score or estimate is made here.
Focus0/100
Scores 0 on concentration: the portfolio splits evenly (0.5/0.5) between "speech recognition / natural language processing" and "speech processing / affective computing (NLP)". Topic counts still cluster on speech — Speech Recognition and Synthesis (232), Speech and Audio Processing (210), Speech and dialogue systems (128) — beside smaller NLP-side clusters in Topic Modeling (56) and Natural Language Processing Techniques (50); by this metric, breadth dominates.
Collaboration53/100
Scores 53. The recent window lists 13 distinct coauthors, and the recurring-partner list centers on speech/NLP colleagues: Esaú Villatoro-Tello, Hritika Sharma, Sergio Burdisso, Alessandra Pinto, Manjunath K E, Hyunggu Jung, Somang Nam, Hasindri Watawana, Thibault Bañeras-Roux and Kadri Hacioglu. A mid-range score suggests a stable, moderately sized network; the two reviewed 2026 papers are middle-authored rather than solo or corresponding work.
Activity100/100
Scores 100, the maximum. Annual work counts stand at 28 (2024), 20 (2025) and 31 (2026), with 79 works in the most recent three-year window; earlier peaks include 30 (2021), 25 (2022) and 22 (2012), and the record reaches back to a single 1997 entry. Sustained, currently near-peak productivity is the clearest signal in this data.
Reproducibility55/100
Scores 55, the mean across 2 papers reviewed for reproducibility. Both Evolve-reviewed 2026 papers rate medium readiness: the HATS-en evaluation study is judged "credible but filtered", and the speech-emotion study reports consistent small gains on IEMOCAP but a single corpus and no significance tests — concrete, stated limits on how readily each pipeline can be re-run.
Research summary
Petr Motlicek is a researcher at Brno University of Technology, Czech Republic (ORCID 0000-0001-6467-1119), working at the intersection of speech technology and natural language processing. The OpenAlex-derived record lists an h-index of 27 and an i10-index of 71, with works indexed continuously from 1997 through 2026. Recent productivity is the standout signal: 79 works in the most recent three-year window, with annual counts of 20 (2025) and 31 (2026) against a prior peak of 30 (2021) — the basis for an activity score of 100. Topically, output clusters on speech: Speech Recognition and Synthesis (232 works), Speech and Audio Processing (210), and Speech and dialogue systems (128), with smaller clusters in Topic Modeling (56) and Natural Language Processing Techniques (50), and an even 0.5/0.5 split between the profile's two assigned fields. Evolve has reviewed two 2026 arXiv papers on which he is a middle, non-corresponding author: a human-aligned study of ASR evaluation metrics beyond WER (introducing the HATS-en preference set) and a discriminative-adaptation approach to speech emotion recognition on IEMOCAP using frozen SpeechLLM hidden states. Both are rated medium reproducibility readiness, giving a reproducibility mean of 55 across 2 papers. Collaboration scores 53, with 13 distinct recent coauthors.
Collaboration pattern
Collaboration is mid-range (score 53) with 13 distinct recent coauthors. The recurring-partner list centers on speech/NLP colleagues — Esaú Villatoro-Tello, Sergio Burdisso, Hritika Sharma, Alessandra Pinto, Manjunath K E, Hyunggu Jung and Kadri Hacioglu. Note two artifacts in that list: an accented and unaccented spelling of Villatoro-Tello, and the researcher's own name, both treated as data artifacts. Both reviewed 2026 papers position him as a middle, non-corresponding author.
Research keywords
Highlights
Human-aligned ASR evaluation: HATS-en and the limits of WER
Reading emotions in token space: discriminative adaptation of SpeechLLMs
Timeline
- 1997First year in the OpenAlex counts-by-year series, with 1 work indexed.
- 2012Annual output reaches 22 works, the highest yearly count before 2021.
- 2021Annual output peaks at 30 works in the counts-by-year series.
- 202631 works indexed so far; two arXiv papers enter the Evolve catalog — HATS-en ASR evaluation and SpeechLLM-based emotion recognition.
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 (h-index 27, i10-index 71, counts-by-year) come from OpenAlex only; no other citation source is included.
- Output and impact radar dimensions are not computed in this snapshot and are reported as missing, not as zero.
- Only two 2026 arXiv papers (both middle-authored, non-corresponding) have been reviewed by Evolve; the reproducibility mean of 55 rests on those 2 papers alone.
- Identity is name-resolved from public bibliographic data (OpenAlex; ORCID 0000-0001-6467-1119); the frequent-coauthor list contains variant spellings and the profile's own name, treated as artifacts.