Crowdsourcing strong labels for sound event detection

Tutkimustuotos: KonferenssiartikkeliScientificvertaisarvioitu

1 Lataukset (Pure)

Abstrakti

Strong labels are a necessity for evaluation of sound event detection methods, but often scarcely available due to the high resources required by the annotation task. We present a method for estimating strong labels using crowdsourced weak labels, through a process that divides the annotation task into simple unit tasks. Based on estimations of annotators' competence, aggregation and processing of the weak labels results in a set of objective strong labels. The experiment uses synthetic audio in order to verify the quality of the resulting annotations through comparison with ground truth. The proposed method produces labels with high precision, though not all event instances are recalled. Detection metrics comparing the produced annotations with the ground truth show 80% F-score in 1 s segments, and up to 89.5% intersection-based F1-score calculated according to the polyphonic sound detection score metrics.
AlkuperäiskieliEnglanti
Otsikko2021 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)
Sivumäärä5
ISBN (elektroninen)978-1-6654-4870-3
DOI - pysyväislinkit
TilaJulkaistu - 13 jouluk. 2021
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE Workshop on Applications of Signal Processing to Audio and Acoustics - , Yhdysvallat
Kesto: 17 lokak. 202120 lokak. 2021

Julkaisusarja

Nimi
ISSN (elektroninen)1947-1629

Conference

ConferenceIEEE Workshop on Applications of Signal Processing to Audio and Acoustics
Maa/AlueYhdysvallat
Ajanjakso17/10/2120/10/21

Julkaisufoorumi-taso

  • Jufo-taso 1

Sormenjälki

Sukella tutkimusaiheisiin 'Crowdsourcing strong labels for sound event detection'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä