Chidzidzo cheAI yeRuzha

Mel-Frequency Cepstral Coefficients

Mel-Frequency Cepstral Coefficients (MFCCs) iboka renhamba rinopfupikisa chimiro cheruzha rwe frequency spectrum semaonerwo azvinoitwa nenzeve dzevanhu.

Teerera kugwaro iriAI-yakarondedzerwa · ~2 min

Pa peji ino2 min verenga
  1. Pfupiso
  2. Kudzika Kwakadzika
  3. Kuitwa muNyika yeChokwadi
  4. Ramangwana reMel-Frequency Cepstral Coefficients
  5. Kuzvishandisa
  6. Ramba Uchiongorora
  7. Mibvunzo inowanzo bvunzwa

Pfupiso

Kwemakumi emakore vaive bhiza rekuzivikanwa kwekutaura, kuzivikanwa kwemutauri, uye kuongorora mimhanzi.

Kudzika Kwakadzika

MFCCs inoshandura chidimbu chipfupi cheodhiyo kuita ingangoita nhamba gumi nenhatu dzinotora timbre yayo. Iyo pombi inotora waveform, yoityora kuita ~ 25ms mafuremu, inokokorodza simba remagetsi kuburikidza neshanduko yeFourier, yobva yatenderedza frequency axis pachiyero chemel, inoisa mabhendi nenzira inoita iyo cochlea: zvakatsetseka pazasi pe1kHz uye zvakakasharara pamusoro. Iwo mel masimba anodzvanywa-akamisikidzwa (kutevedzera nzwisiso yekuziva) uye pakupedzisira akapfuura nemu discrete cosine shanduko, inovashongedza uye inoisa ruzivo mune ekutanga coefficients. Mhedzisiro yacho yakasimba kune ruzha uye kukwirira kwemutauri, ndosaka classic Yakavanzwa Markov Model uye Gaussian Mixture Model matauriro masisitimu aivimba neMFCCs angangoita pasi rose asati adzidza zvakadzama.

Kuitwa muNyika yeChokwadi

Acoustic maficha echinyakare HMM-GMM matauriro ekuziva senge yekutanga Sphinx uye HTK masisitimu

Kusimbisa mutauri uye diarization, kusiyanisa kuti ndiani ari kutaura parunhare

Mimhanzi genre classification uye rwiyo kudhinda zvigunwe (Shazam-style timbre matching)

Kuona kukanganisa kwemuchina kana kufona kwemhuka kubva kuodhiyo mumaindasitiri uye bioacoustic yekutarisa

Ramangwana reMel-Frequency Cepstral Coefficients

Kupera-kusvika-kumagumo akadzika network anowedzera kudzidza maficha akatwasuka kubva kune akaomeswa waveforms kana log-mel spectrograms, kusvetuka iyo DCT, saka yakachena MFCCs iri kupera kubva kumamiriro-e-iyo-ASR. Zvakadaro ivo vanoramba vakakurumbira kune akareruka, pane-mudziyo, uye akaderera-data mabasa: kiyi yekuona, kuona chiitiko chezwi, kudhindisa zvigunwe zvekuteerera, uye bioacoustics. Tarisira kuti maMFCC arambe achienderera mberi seanoshanda, anodudzira ekutanga kunyangwe akadzidza kumberi-kunotonga mhando hombe.

Kuzvishandisa

Nei zvichikosha

  • Svika uye svika. Inonatsiridza kusvikika kuburikidza nekunyora, kurondedzera, uye mazwi ekubatanidza.
  • Mutengo uye bhajeti. Zvikwata zveMedia zvinogona kutumira odhiyo yakakwenenzverwa nekukurumidza nemabhajeti madiki.
  • Kumhanya uye chiyero. Masisitimu anotarisana nevatengi anogona kugadzirisa kutaurirana kwekutaura pamwero mukuru.

Chenjera

  • Kushandisa izwi zvisizvo uye njodzi dzekuedzesera dzinowedzera kana chibvumirano chisipo.
  • Kururama kunogona kudonha mumitauro, mataurirwo, kana nharaunda dzine ruzha.
  • Synthetic audio inogona kukanganisa kutaura kwechokwadi isina mavara akajeka.

Matanho anotevera

  1. Wana mvumo yakajeka yekutora inzwi, kugadzira, uye kushandisa zvakare.
  2. Yedza mhando pavatauri vakasiyana uye mamiriro ekumashure.
  3. Tsanangura apo munhu anofanira kuongorora kana kubvumidza zvabuda.
  4. Label synthetic odhiyo uye chengetedza marekodhi ekuzvidavirira.

Ramba Uchiongorora

Tsamba yemahara

Tora iyo yemazuva ese AI pfupiso

Nhatu dzakasimbiswa dzeAI nyaya mangwanani ega ega pakati pevhiki, dzakanyorwa neChirungu chakajeka. Yemahara zvachose, hapana ads.

Imwe email zuva rega rega revhiki. Bvisa kunyoresa kamwe chete. Hatimbotengese kana kugovera kero yako.

Zviedze iwe pachako

Take the Mel-Frequency Cepstral Coefficients quiz

Mhinduro yepakarepo pamhinduro yega yega, uye chitupa chinogovaniswa chine ID inobatika kana wapedza kosi.

Tanga mibvunzo

Tsigira dzidzo yemahara yeAI. AI Understanding ndeye 501(c)(3) isingabatsiri - hapana ads, hapana paywall, zvachose. Ita chipo

Mibvunzo inowanzo bvunzwa

Chii chinonzi Mel-Frequency Cepstral Coefficients?

Mel-Frequency Cepstral Coefficients (MFCCs) iboka renhamba rinopfupikisa chimiro cheruzha rwe frequency spectrum semaonerwo azvinoitwa nenzeve dzevanhu. Kwemakumi emakore vaive bhiza rekuzivikanwa kwekutaura, kuzivikanwa kwemutauri, uye kuongorora mimhanzi.

Ko iyo 'mel' muMFCC inorevei?

Iyo mel scale warps frequency kuitira kuti nhanho dzakaenzana dzinzwike zvakaenzana kure kure kuvanhu, dzakapatsanurwa zvakadzika pamhepo dzakaderera uye dzakakora kune dzakakwirira.

Ndeipi shanduko inoshandiswa kwekupedzisira kugadzira cepstral coefficients?

Mushure mekunge log-mel simba raverengerwa, discrete cosine inoshandura inoishongedza uye inorongedza ruzivo mune mashoma ekutanga coefficients.

Sei maMFCC akasimba kusvika pakutaura kwemutauri?

Low cepstral coefficients inobata iyo vocal-tract envelope (phonetic content) nepo yakakwirira ichitora kukwirira, saka kuchengeta iyo yakaderera de-kusimbisa kukwirira.

Anenge mangani maMFCC coefficients anowanzo chengetwa pafuremu?

Sarudzo yakajairika ndeye gumi nematatu coefficients, dzimwe nguva inowedzerwa neiyo deltas, iyo inopfupikisa timbre.

Sei danda richishandiswa kune mel-band simba?

Maonero eruzha rwemunhu angangoita logarithmic, saka kudzvanywa kwelogi kunoita kuti maficha ari nani aenzanise maonero uye anodzikamisa dhizaini.