Xeetu seen akustik
Classification scene akustik (ASC) dafay tàggat masin yi ñu xàmmee environmaa bi ñu defee enregistrement bi, mbedd bu bari nit, park bu dal, saxaar, kafe, ci son kese.
Résumé
It gives devices a sense of 'where they are' using audio alone.
Plongeur bu xóot
ASC dafay laaj benn model mu jox benn clip audio lëmm ci benn etiketu seen ci texture son bi yépp, du benn xew-xew. ASC wuute na ak gis-gis yiy xew ci son, ndax dafay gis xaj buy waay wala sirene, ASC dafay àtte njaxasu ambiant bi, hum bi, reverberation bi, ak dëgëraay bi ci son yiy jaxasoo. Sistem yi dañuy soppi audio yi ci spectrogram log-mel ba noppi ñu jox leen CNN yi wala transformatëri audio yi, ñuy faral di jëfandikoo yokk done yu melni mixup ak SpecAugment ngir xeex overfitting ci done yu néew. The annual DCASE Challenge has driven progress, especially on hard problems like device mismatch (a model trained on one phone's microphone failing on another) and building tiny, low-power models that run on edge devices.
Gis-gis xarala
Jafe-jafe bi gëna mag mooy seen yi dañu leen di fësal ci lim yu yàgg yi, du ci xew-xew yu gàtt, moo tax model yi dañuy boole màndarga yi ci segond yu bari. Ngir mëna dundu ci yeneen aparey enregistrement, ingénieur yi dañuy jëfandikoo pexe yu méngoo ak domen bi ak yokk xam-xam aparey biy nuru tontu fréquence mikro yi. Sistem DCASE yu bari yuy jël ndam li dañuy xayma ak dagg seen reso ngir mëna daje ak budget memory bu tar (dafay faral di nekk ci suufu 128 KB), loolu dafay firndeel ni ASC mën na dox ci aparey te du am cloud processing.
njeextalu pexe
Dugg ak yegg
Dafay gëna yombal jëfandikoo gi jaaraleko ci transkripsioŋ, nettali ak interfaasu baat.
Njëgg ak budget
Ekipu mejaa yi mën nañu yónnee audio bu leer ci anam wu gëna gaaw te seen xaalis gëna néew.
Gaawaay ak yaatuwaay
Sistem yiy jàkkarloo ak kiliyaan bi mën nañu def waxtaan ci anam wu gëna yaatu.
Ëlëgu xaajale seen akustik
ASC mingi nekk luñuy tabax ci aparey yiy xam contexte: aparey yiy déglu yuy méngoo seen bopp ak resto, telefon yuy soppi profil soo duggee ci oto, ak kër yu xarañ yuy xam liggéey te amul kamera (baña yàq sa bopp). Gëstu dafay puus ngir mëna ànd ak environmaa bu bees, dëgër ci bépp mikro, ak model yu xarañ lool. Buñu ko boole ak gis-gis xew-xew yi am son, ASC dina may masin yi ñu gëna riis, di wéy di xam li leen wër.
Doxal ci àdduna dëgg
Aparey auditif yi dañuy xàmmee resto bu bari bruit ak néeg bu dal ba noppi di wàññi bruit bi ci saasi
Telefoon yu xarañ yi di soppi ci profil 'dawal' wala 'bitti' bu sukkandiko ci son biy wër
Sistemu kër yu xarañ yiy baña yàq mbir yu nëbbu, di jëlee audio bi ci wideo bi, te du ci wideo bi
Jumtukaayi enregistrement ak bioakustik yuy tànnal waxtu enregistrement yi ci xeetu dëkkuwaay
Risk yi ak balustrade yi
Jëfandikoo baat ci anam wu jaarul yoon ak niru ak nit dafay gëna yokk sudee nanguwul.
Jaar-jaar mën na wàññeeku ci aksan yi, dialect yi wala barab yu bari xumbaay.
Audio synthetik mën nañu ko jaawale ak wax ju dëggu sudee amul etiket bu leer.
Roadmap ngir samp gi
Wutal ndigal bu leer ngir jàpp baat bi, klone ko ak jëfandikoowaat ko.
Saytu kalite ci kàddukat yu bari ak anam yu bari ci ginaaw.
Mandargal kañ la nit wara xoolaat wala nangu ay génne.
Etiketu audio synthetik te nga denc dokimaa ci fimu bawoo ngir mëna lim.
Weyal di banneexu
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Acoustic Scene Classification quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Gis bi ci topp
Xeetu jamonoy lëkkaloo
Laaj yi ñuy faral di laaj
What is Acoustic Scene Classification?
Classification scene akustik (ASC) dafay tàggat masin yi ñu xàmmee environmaa bi ñu defee enregistrement bi, mbedd bu bari nit, park bu dal, saxaar, kafe, ci son kese. Dafay may aparey yi ñu xam 'fu ñu nekk' suñu jëfandikoo audio kese.
Lan moo wuutale xaajalug seen akustik ak gis xew-xewu son?
ASC dafay màndargaal seen bi yépp (lu melni, 'mbedd', 'park'), waaye gis-gis yi am ci son yi dañuy gis son yi melni sirene wala bark.
Luy jafe-jafe 'jamonowul aparey' ci ASC?
Mikro yu bari dañu am tontu yu wuute ci frequency, moo tax benn model buñ tàggat ci benn aparey dafay faral di yàqu ci enregistrement yu bawoo ci beneen aparey, muy jafe-jafe bu mag ci ASC.
Ban misaalu dugal mooy standard ci xeetu ASC yi?
ASC ni IA audio yu bari, dafay soppi clip yi ci spectrogram log-mel yi CNN yi wala transformatër yi mëna def.
Lan moo waral model ASC yi di boole ay man-man ci ay segond yu bari te baña boole ay saa yu benn?
Liy wane benn seen mingi aju ci texture bi ak ambiance bi ci diir bi, moo tax model yi dañuy boole xibaar yi ci clip bi yépp.
Ban jafe-jafe gëstu moo jur yokkute yu bari ci ASC?
At mu nekk DCASE (Gis ak xaaj ay seen ak xew-xew akustik) Challenge mooy référence bi gëna am solo biy puus ASC ci kanam.