HejuruUbuyobozi bukurikira
Gutegura Inyandiko-Kuri-Imvugo Dataset
Tekiniki
Ubuyobozi bwa tekiniki
Offline inyandiko-ku-ijambo ihindura inyandiko yanditse mu mvugo ukoresheje software yakuweho na moderi cyangwa dosiye zijwi.
Umva iki gitaboAI-byavuzwe · ~3 min
Piper na Kokoro ni imishinga itandukanye ifite ibishushanyo mbonera bitandukanye, gupakira, gukwirakwiza ururimi, hamwe nibyuma bikenerwa, gereranya rero ibyangombwa hamwe nimpushya zo gukora neza nijwi uteganya gukoresha.
Umwandiko-ku-mvugo, cyangwa TTS, ikarita yanditse inyandiko yanditswemo imvugo. Gukoresha inzira kumurongo bivuze ko porogaramu ifite igihe gikenewe cyo gukora na moderi cyangwa dosiye yijwi kubikoresho byaho kandi ntibikeneye kohereza buri jambo kuri serivise yakiriwe na TTS. Itezimbere kuboneka ariko ihererekanya kwishyiriraho, kubika, ninshingano zikorwa kubakoresha. Piper ni umushinga ufungura isoko ya TTS hamwe namajwi ashobora gukururwa hamwe nibikoresho byaho. Umushinga wa Piper uriho ubu ubungabunzwe mububiko bwa OHF-Ijwi kandi uruhushya rwarwo hamwe nibisobanuro birambuye bigomba kugenzurwa aho. Kokoro numuryango wicyitegererezo wa TTS hamwe namakarita yacyo yicyitegererezo, amajwi, amasomero yerekana, hamwe namakuru yindimi. Amazina ntabwo yerekana icyitegererezo cyububiko, ubwiza, cyangwa ibidukikije bishyigikiwe. Guhindura no gupakira impinduka zirashobora kubaho, koresha rero ibyangombwa byumushinga aho gukoresha inyigisho ishaje. Icyitegererezo cyijwi ntabwo arikintu kimwe nigihe cyo kugitwara. Imiterere yicyitegererezo, fonemizers, inkoranyamagambo yamagambo, udupapuro twindimi, hamwe n’amajwi biterwa byose bigira ingaruka niba synthesis ikora. Moderi yoroheje irashobora guhuza igikoresho cya CPU kibujijwe, mugihe ikindi gishobora gusaba kwibuka cyane cyangwa kwihuta. Gupima umuvuduko wo kubyara, igiciro cyo gutangira, kwibuka ibirenge, hamwe nubuziranenge bwamajwi kubikoresho bigenewe intego. Kwiyongera cyangwa gutandukana kwinyuma bishobora guhindura imikorere nibisohoka. Ururimi nijwi bigomba kugenzurwa kurwego rwicyitegererezo. Igihe cyo gushyigikira ururimi ntabwo bivuze ko ijwi ryose rishyigikiye neza. Amazina akwiye, amagambo ahinnye, imibare, utumenyetso, hamwe na code-guhinduranya bishobora gusaba inyandiko isanzwe cyangwa amategeko yo kuvuga. Gutegera ibizamini hamwe nibice byerekana birasobanutse kuruta guca urubanza interuro imwe. Uruhushya no kwemererwa bikeneye cheque zitandukanye. Kode yumushinga, uburemere bwikitegererezo, hamwe nijwi ryijwi ryumuntu rishobora kugira amagambo atandukanye. Niba ushushanya ijwi ryumuntu wamenyekanye, shaka uruhushya kandi ukurikize amategeko akurikizwa. Kurangiza kumurongo byonyine ntabwo byerekana ko gukwirakwiza cyangwa gukoresha ubucuruzi byemewe. Inkomoko yinyandiko, verisiyo, hamwe nijwi ryamajwi mbere yo kohereza.
Raspberry Pi iratangaza sensor status ikoresheje ijwi rya Piper ryakuweho nyuma yo gupima ubukererwe bwibisekuru hamwe nibuka kubikoresho.
Porogaramu ya desktop ikoresha moderi ya Kokoro mugace kandi igenzura imvugo ishyigikiwe nijwi mbere yo kubyara indimi nyinshi.
Itsinda ryibicuruzwa risubiramo uruhushya rwigihe cya TTS kandi rugenzura ukundi amagambo yometse kuri buri jwi cyangwa moderi yakuweho.
Umukoresha-ureba-umusomyi agabanya inyandiko ndende mo ibice, abika utumenyetso, kandi agenzura ubudahwema bwamajwi kumipaka.
Offline TTS birashoboka ko izakomeza gutera imbere nkuko ibirenge byintangarugero bigabanuka kandi igihe cyo gukora gishyigikira ubwoko bwibikoresho byinshi. Abakoresha barashobora kubona imvugo yagutse hamwe nijwi ryamajwi, nubwo ubwiza nubwiza bizakomeza kutaringaniza kurugero. Gupakira byoroshye bishobora koroshya kwishyira hamwe, mugihe icyitegererezo gisobanutse hamwe nijwi ryijwi ryafasha amakipe gusuzuma imiterere yimikoreshereze. Synthesis yaho iracyakenera ibizamini byo gutega amatwi, gupima ibyuma, gusubiramo impushya, no gucunga neza indangamuntu. Ibipimo byibikoresho bigomba gushyiramo inyandiko ifatika kandi ubukonje butangira. Reba ibi bintu mbere yo koherezwa mugari.
Ibaruwa y'amakuru y'ubuntu
Inkuru eshatu zizewe za AI buri gitondo cy'iminsi y'akazi, zanditse mu rurimi rworoshye. Ubuntu iteka ryose, nta kwamamaza.
Imeri imwe buri munsi y'akazi. Kwiyimura mu buryo bumwe. Ntitwigera tugurisha cyangwa ngo dusangire aderesi yawe.
Isuzume
Ibisubizo ako kanya kuri buri kibazo, n'icyemezo gishobora gusangirwa gifite ID yagenzurwa umaze gutsinda isomo.
Fasha uburezi bwa AI bw'ubuntu. AI Understanding ni umuryango utegamiye kuri leta wa 501(c)(3) — nta kwamamaza, nta kwishyura, iteka ryose. Tanga impano
Offline inyandiko-y-ijambo ihindura inyandiko yanditse mu mvugo ukoresheje software yakuweho na moderi cyangwa dosiye zijwi. Piper na Kokoro ni imishinga itandukanye ifite ibishushanyo mbonera bitandukanye, gupakira, gukwirakwiza ururimi, hamwe nibyuma bikenerwa, gereranya rero ibyangombwa hamwe nimpushya zo gukora neza nijwi uteganya gukoresha.
TTS yaho ikomatanya amajwi yerekana amajwi yanditse kubikoresho.
Inyandiko zabo nibikorwa byabo biratandukanye, kubwibyo ubushobozi ntibukwiye gufatwa nkibishobora guhinduka.
Kode, uburemere hamwe namakuru yijwi arashobora gutwara amategeko atandukanye.
Igihe-nyacyo kigabanya synthesis igihe cyakozwe amajwi; kwibuka no gukonja-gutangira gutinda nibindi bipimo.
Umwandiko usanzwe hamwe nimbibi zishobora kugira ingaruka zifatika.
Komeza wige
Abandi bayobozi batoranijwe kuriyi ngingo
HejuruUbuyobozi bukurikira
Gutegura Inyandiko-Kuri-Imvugo Dataset
Tekiniki