Audio AI JAGORA

OpenAI Waswasi

Whisper shine tsarin gano magana ta atomatik na OpenAI wanda ke fassara da fassara sautin da ake magana a cikin harsuna da dama.

2 min karatuAn sabunta ta ƙarshe

Dubawa

It matters because it brought robust, free, near-human transcription to anyone who can run the model.

Zurfafa nutsewa

An sake shi a cikin Satumba 2022, an horar da Whisper akan kusan sa'o'i 680,000 na yaruka da yawa, sautin ayyuka da yawa da aka tattara daga gidan yanar gizo. Wannan babbar maɓalli daban-daban shine sirrin ƙarfinsa: yana sarrafa lafazin, hayaniyar baya, da jargon fasaha fiye da tsofaffin tsarin, ba tare da buƙatar daidaitawa ga kowane sabon yanki ba. Wasiwasi na iya kwafin magana a cikin yaren asali, fassara magana daga harsuna da yawa zuwa Turanci, gano harshen da ake magana, da ƙara tambarin lokaci. OpenAI yana fitar da ma'aunin ƙima da lambar a bayyane, don haka yana aiki a cikin gida akan kwamfutar tafi-da-gidanka ko a cikin cibiyar bayanai, wanda ya haifar da fashewar kayan aikin al'umma, sake aiwatarwa cikin sauri, da aikace-aikacen da aka gina a samansa. Daidaito ya bambanta ta harshe da ingancin sauti, kuma kamar duk irin waɗannan tsarin yana iya 'ra'ayi' rubutu lokaci-lokaci.

Fahimtar Fasaha

Waswasi shine mai rikodin rikodin canza canji wanda aka horar dashi azaman jerin-zuwa-jeri aiki. Ana jujjuya sauti zuwa sikirin log-Mel, wakilcin gani-kamar na mitoci akan lokaci, wanda ke aiwatar da encoder. Mai yankewa sai yayi tsinkaya alamomin rubutu, sharadi ta hanyar alamu na musamman waɗanda ke gaya wa ƙirar aikin da zai yi: kwafi, fassara, gano harshe, ko ƙara tambarin lokaci. Saboda an koya daga sautin gidan yanar gizo mai rauni mai rauni a cikin ayyuka da yawa lokaci guda, ƙirar ƙira ɗaya ta ke ba da juzu'i a maimakon ana sauraren maƙasudin maƙasudi ɗaya.

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 OpenAI Waswasi

Whisper ya zama tsohuwar tubalin ginin rubutu, kuma yanayin yana zuwa ga sauri, ƙarami, da bambance-bambancen lokaci na gaske waɗanda ke gudana akan wayoyi da na'urorin gefen. Yi tsammanin tallafin yawo mai ƙarfi, mafi kyawun rarrabuwar magana, da haɗin kai tare da manyan samfuran harshe don tsaftacewa, taƙaitawa, da taken kai tsaye. Bude ma'aunin nauyi yana nufin al'umma suna ci gaba da inganta shi, yayin da OpenAI da wasu ke tura sabbin nau'ikan magana. Rage rubutun halat, musamman a cikin amfani da magani da na doka, ya kasance babban fifiko.

Aiwatar da Gaskiyar Duniya

Dan jarida yana rubuta tambayoyin da aka yi rikodin ta atomatik maimakon buga su da hannu

Dandali na podcast yana haifar da rubuce-rubucen da za'a iya bincikawa da rubutun kalmomi don kowane labari

Kayan aiki na taro yana samar da taken kai tsaye da rubutaccen rikodin kiran bidiyo

Wani mai bincike yana fassara rikodin filin magana zuwa rubutun Turanci don bincike

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 OpenAI Whisper 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

Daidaita Kalma Mai Tafiyar Lokaci

Tambayoyin da ake yawan yi

What is OpenAI Whisper?

Whisper shine tsarin gano magana ta atomatik na OpenAI wanda ke fassara da fassara sautin da ake magana a cikin harsuna da dama. Yana da mahimmanci saboda ya kawo ƙaƙƙarfan, kyauta, kwafin ɗan adam ga duk wanda zai iya tafiyar da ƙirar.

Menene babban makasudin OpenAI's waswasi?

Waswasi shine tsarin gane magana ta atomatik wanda ke yin kwafi da fassara sautin magana a cikin yaruka da yawa.

Kusan nawa ne aka koya wa Whisper a kai?

An horar da Whisper akan kusan sa'o'i 680,000 na harsuna da yawa, sauti mai yawan aiki da aka tattara daga gidan yanar gizon, wanda ke motsa ƙarfinsa.

Wane wakilcin odiyo ke aiwatar da ɓoyayyen ɓoyayyiyar Whisper?

Ana jujjuya sauti zuwa sikirin log-Mel, wakilcin abun ciki na mitar a kan lokaci, wanda mai canza canji ya karanta.

Wanne iyawa ne Whisper BAYA bayarwa na asali?

Whisper yana sarrafa rubutun, fassara zuwa Turanci, gano harshe, da tambura, amma ba janareta na bidiyo ba.

Me yasa Whisper ya haifar da guguwar kayan aikin al'umma da ƙa'idodi?

Saboda OpenAI sun bude ma'auni da lambar, masu haɓakawa za su iya gudanar da Whisper a cikin gida kuma su gina kayan aiki da yawa da saurin aiwatarwa.