Se-hindi e mahala

Ngola Se-hindi audio le video ho ea ho mongolo ka AI. E potlakile, e nepahetse, e mahala.

Ho sebetsa joang

  1. E-ba le Free.ai Transcriber
  2. Kopitsa faele ea hau ea Se-hindi ea audio kapa video
  3. AI ea rona e tla fumana Se-hindi ka ho toba'me ngole
  4. Kopitsa transcript ea hau e le tekanyo kapa lihlooho tsa SRT

Se-hindi Likarolo tsa ho ngola

  • E sebetsa ka faster-whisper (MIT licensed)
  • Ho fumana puo Se-hindi ka ho toba
  • E tšehetsa MP3, WAV, MP4, M4A, FLAC, le tse ling
  • Timestamps le ho romelloa ha lihlooho (SRT)
  • Ha ho na liphelelo tsa boholo ba faele ka likhetho tse lefelloeng
  • Secha le se sireletsehileng -- lifaele li tlosoa ka mor'a ho sebetsana

Lintlha tsa puo

SenyesemaneSe-hindi
ISO Codehi
Mofuta oa AIfaster-whisper
Theko_Tsela

Li-languages tse ling

Tseba Li-Languages Tseo

Lipotso tse botsoang khafetsa

Whisper large-v3-turbo e sebetsana le Se-hindi ka ho feletseng - 7-15% palo ea liphoso tsa mantsoe ka molumo oa benchmark. E-ba le liphetoho tse sa tloaelehang ka li-entities tse bitsoang, lipalo, le mantsoe a tekheniki a thata; karolo e kholo ea transcript e tla ba ea nepahetseng.(Tier B, 7-15% word error rate ka lihlopha tsa benchmark - re phatlalatsa li-tiers tsa WER tse tšepahalang ho feta li-claims tsa thekiso.)

E-na — Se-hindi transcription e nka ho tloha ho pholletsa le letsatsi token pool ea hao mahala pele. Audio theko ka 50 tokens ka metsotsoana, ka hona pholletsa le letsatsi pool Anonymous e ka fihlela lihora tse'maloa tsa audio ka letsatsi. Signed-ka akhaonteng fumana pool e kholo le 10,000 signup tokens. Past hore, $ 1 reka 750,000 tokens (~ 250 lihora tsa audio).

Hindi audio hangata e kopanya le Senyesemane (Hinglish) ka puo ea toropong. Whisper e sebetsana le ho kopanya le ho ngola mantsoe a Senyesemane ka ho ngola ka Latin le mantsoe a Hindi ka Devanagari ka har'a transcript e le 'ngoe. Ho bua ka litoropo le mantsoe a litoropo a boima ho ka ba le ho nepahala ha tier-C.

MP3, WAV, M4A, FLAC, OGG, OPUS, le WEBM li amoheloa ka kotloloho. Bakeng sa video (MP4, MOV, MKV) re tlosa li-track tsa audio ka lehlakoreng la mosebelisi pele re li romella ho Whisper — ha o hloka ho fetola ntho efe kapa efe ka boeona. Pipeline e ts'oanang ntle le puo ea mohloli, ho kenyeletsoa Se-hindi.

Li-account tse kentsoeng ka har'a li-account li fihla ho 2 GB. Nako e telele ha e lekanyetsoe - li-file tse telele li kopanngoa ka ho toba (lifensetere tsa metsotsoana e 30 le ho koaloa)'me li kopanngoa ho ea ho transcript e le' ngoe le li-timestamps tse tsoelang pele. Li-recording tsa Se-hindi tse fetang lihora tse ngata (podcasts, lihlooho tse felletseng, liboka) li sebetsa hantle.

Ha ho joalo — ho arolelana lingoliloeng tsa moqoqo ho hokahantsoe ka ho sa feleng bakeng sa Se-hindi transcript e ngoe le e ngoe. Liphetho li arotsoe e le Moqoqo 1 / Moqoqo 2 / Moqoqo 3 le li-timestamps, ka hona lipotso tsa puisano, li-panel discussions, le liketsahalo tse ngata li khutlela ka letšoao. Ho arolelana lingoliloeng tsa moqoqo ho sebetsa ka mokhoa o fapaneng'me ho sebetsa ka tsela e ts'oanang ka lipuo tsohle tseo re li tšehetsang.

E-ea — kopanya URL ho /transcribe/youtube/ bakeng sa YouTube kapa /transcribe/podcast/ bakeng sa podcast feeds (Apple, Spotify, RSS). Re kenya audio, e tsamaisa ka Whisper le language=hi,'me re khutlisetse transcript le timestamps le li-labels tsa moqolotsi oa litaba. Lintlha tsa Se-hindi: Litlhaku tsa puisano ea WhatsApp, livideo tsa YouTube le livideo tse khuts'oane ke li-workloads tse tloaelehileng ka ho fetisisa tsa Se-hindi — kenya URL ho /transcribe/youtube/ kapa u romelle audio ka kotloloho.

Whisper e theko e boima ka li-token tsa 50 ka metsotsoana ea audio, ka hona ho ngolisoa ha hora e le 'ngoe ke li-token tsa ~ 3,000. $ 1 e reka li-token tsa 750,000, tse sebetsang ho fihlela lihora tse 250 tsa audio ka dollar. Basebelisi ba bangata ha ba sebelise ntho e ngoe le e ngoe - pool ea letsatsi le letsatsi e nang le li-clip tse khuts'oane, li-notes tsa puisano le podcasts tse le' ngoe.

Ee — li-timestamps tsa lefatše la sehlooho (ka ~10-30 sekontiri) le tsa lefatše la mantsoe li fumaneha. Lefatše la mantsoe ke la morao-rao bakeng sa ho romelloa ha li-subtitles tsa VTT/SRT ka hona li-captions li hokahana line-by-line. Ka API beha timestamps="word" ka'mele oa kopo. Se-hindi transcripts li khutlisoa ka Devanagari script (UTF-8).

Ee. POST audio (multipart/form-data, field name "file") ho /v1/transcribe/ le language=hi — kapa u se ke ua beha parameter ea puo ho etsa hore Whisper e fumane ka ho toba. E khutlisa JSON le transcript, segments, timestamps, le labels tsa motsamaisi. Litlhahlobo tse felletseng le li-snippets tsa SDK li ka fumaneha ho /api/.

E-ea — ha transcription e felile, tobetsa ho fetolela kapa ho kopanya tekanyo ho /translate/. Se-hindi e kopantsoe le lipuo tsohle tse ling tseo re li tšepang (200+). Bakeng sa lihora tsa kopano, transcript e fetisoa ka /summarize/; bakeng sa ho ngola ka letsoho, e romelle ho /voice/tts/ ho etsa molumo ka puo e loketseng.

Whisper e koetlisitsoe ka lihora tse likete tse likete tsa molumo oa lefatše la 'nete, ka hona e tolera molumo oa morao-rao le ho ngolisoa ka boleng ba mohala ho Se-hindi. Bakeng sa liphetho tse ntle, fana ka molumo o hloekileng (mikrofono ea headset, ha ho na lebokose la molumo) — ka taelo ena molumo o eketsa palo ea liphoso tsa baseline.Haeba transcript e sa khone ho sebelisoa, romella lengolo-tsoibila contact@free.ai le faele — re tla lefa tokens'me re sheba hore na e mong oa li-engine tse fapaneng o sebetsana joang le audio ea hau ka ho fetisisa.

U rata Free.ai? Reka le metsoalle ea hau!

Ratela leqepheng lena