Audio AI JAGORA

Fasalolin Filterbank da PLP

Fasalolin Filterbank da Perceptual Linear Prediction (PLP) hanyoyi ne na taƙaita siginar magana cikin ƙanƙanta, lambobi masu ma'ana waɗanda ƙirar koyon injin za su iya amfani da su.

2 min karatuAn sabunta ta ƙarshe

Dubawa

They matter because they let speech recognizers focus on the parts of sound humans actually hear, discarding irrelevant detail.

Zurfafa nutsewa

Don juyar da ɗanyen sautin sauti zuwa fasali, ana raba siginar zuwa gajerun firam kuma an wuce ta cikin banki na matattara masu ruɗewa da ke sararin sama a ma'aunin mel, wanda ke kwaikwayi hankalin mitar kunne mara sahihanci. Ƙirƙirar makamashi a cikin kowane tacewa yana samar da fasalulluka na log-mel filterbank, mafi rinjayen shigarwa don ƙirar magana mai zurfi na zamani. PLP, wanda Hynek Hermansky ya ƙera, yana ƙara ƙarin ilimin halin ɗan adam: yana amfani da madaidaitan madauri mai girman haushi, mitoci masu nauyi daidai da ƙara kamar yadda kunne ke yi, da matsi-tushen ƙarfi-zuwa sauti, sannan ya dace da ƙirar sandal (daidaitacce) samfurin don daidaita bakan. Sakamakon shine ƙaramin wakilci mai ƙarfi ga mai magana da bambance-bambancen tashoshi. MMFCs ɗan uwan ​​​​kusanci ne wanda ke ƙara canjin cosine don ƙawata abubuwan fitar da bankin tacewa.

Fahimtar Fasaha

Makullin ra'ayin shine warping na fahimta: an sake tsara hertz na layi zuwa mel ko sikelin haushi don haka masu tacewa suna kunkuntar a ƙananan mitoci kuma suna faɗi a manyan, daidai da ƙudurin cochlear. PLP's daidai-ƙarfi preemphasis da kuma tushen cube-tushen matsawa yadda tsinkayar ƙarar kunne ba ta kan layi. Matakin tsinkayar layi na ƙarshe ya dace da ambulaf mai santsi mai santsi, yana ɗaukar siffar muryoyin murya yayin da yake murkushe jituwar farar da ta bambanta tsakanin masu magana.

Dabarun Tasiri

Shiga ku isa

Yana inganta samun dama ta hanyar rubutu, ba da labari, da mu'amalar murya.

Kudin da kasafin kuɗi

Ƙungiyoyin kafofin watsa labaru na iya jigilar sauti mai gogewa cikin sauri tare da ƙaramin kasafin kuɗi.

Gudu da sikelin

Tsarin fuskantar abokin ciniki na iya aiwatar da hulɗar magana a mafi girman ma'auni.

Makomar Filterbank da Fasalolin PLP

Cibiyoyin sadarwar jijiyoyi masu zurfi suna ƙara fifita danyen log-mel filterbanks akan ingantattun kayan aikin PLP ko MMFC, saboda hanyar sadarwar tana koyon nata sauyi fiye da kayan ado na hannu. Ƙarshen gaba za a iya koyo kamar SincNet da wav2vec waɗanda ke aiki akan ɗanyen raƙuman ruwa. Duk da haka, mel filterbanks suna kasancewa a ko'ina a matsayin tsayayye, shigarwa mai rahusa, da ka'idojin fahimta da ke bayan PLP suna ci gaba da sanar da yadda injiniyoyi ke tsarawa da fassara waɗannan wakilcin da aka koya.

Aiwatar da Gaskiyar Duniya

Ƙididdigar fasalulluka 40 log-mel filterbank kowane firam azaman shigarwa zuwa hanyar sadarwa-zuwa rubutu

Amfani da fasalulluka na PLP a cikin amo-ƙarfin tsarin umarnin murya don motoci

Bututun tantance mai magana waɗanda suka dogara da fassarori masu ɓarnatar fahimta

Haɓaka maɓalli akan na'urori masu ƙarancin ƙarfi inda ƙaƙƙarfan fasalulluka na filterbank suna rage ƙididdiga

Hatsari & Tsare-tsare

Rashin amfani da murya da haɗarin kwaikwaya yana ƙaruwa lokacin da aka rasa izini.

Daidaituwa na iya faɗuwa cikin lafuzza, yaruka, ko mahalli masu hayaniya.

Ana iya kuskuren sauti na roba don ingantacciyar magana ba tare da bayyananniyar lakabi ba.

Taswirar Hanya

1

Sami tabbataccen izini don ɗaukar murya, cloning, da sake amfani.

2

Gwajin ingantattun masu magana daban-daban da yanayin baya.

3

Ƙayyade lokacin da dole ne ɗan adam ya duba ko ya amince da abubuwan da aka fitar.

4

Yi lakabin sauti na roba da kuma adana bayanan da aka tabbatar don yin lissafi.

Ci gaba da Bincike

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 Filterbank and PLP Features quiz

Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.

Fara tambayoyi

Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation

Jagora na gaba

Ƙimar Layi na Layi da Daskararrun Ƙimar Siffar

Tambayoyin da ake yawan yi

What is Filterbank and PLP Features?

Fasalolin Filterbank da Perceptual Linear Prediction (PLP) hanyoyi ne na taƙaita siginar magana cikin ƙanƙanta, lambobi masu ma'ana waɗanda ƙirar koyon injin za su iya amfani da su. Suna da mahimmanci saboda suna barin masu gane magana su mai da hankali ga sassan mutane masu sauti a zahiri su ji, suna watsar da cikakkun bayanai marasa mahimmanci.

Me yasa bankunan tace magana aka bazu akan ma'aunin mel ko haushi maimakon layin hertz?

Ma'aunin mel da haushi yana daidaita ƙayyadaddun ƙudurin cochlear ɗan adam, wanda ya fi kyau a ƙananan mitoci da ƙaranci a mitoci masu girma.

Menene matakin tsinkayar layi a cikin PLP ya cim ma?

PLP ya dace da kowane nau'in sandar sanda don samar da ambulan mai santsi, yana ɗaukar halaye na murya yayin da yake murkushe masu jituwa mai dogaro da sauti.

Wanne nau'in fasalin shine mafi girman shigarwa don ƙirar gane magana mai zurfi na zamani?

Fasalolin log-mel filterbank sune daidaitattun bayanai, ƙanƙanta, shigarwar fahimta mai ma'ana don ƙirar magana mai zurfi na yau.

Ta yaya MMFCs suka bambanta da albarkatun log-mel filterbank?

MFCCs suna amfani da canjin cosine mai hankali a saman ƙarfin log-mel filterbank don ƙawata su zuwa ƙananan ƙididdiga.

Wane nau'in fahimta PLP ya haɗa da cewa ainihin mel filterbanks ba sa?

PLP yana ƙara daidai-ƙarfin ƙarar ƙarar ƙarfi da matsi-tushen ƙarfi-zuwa-ƙarfi zuwa mafi kyawun ƙirar tsinkayen ƙarar ɗan adam.