LLM EU
Menu

Whisper large-v3

OpenAI

Non ospitato — backend mock Pesi aperti

Sovranità e hosting

Non ospitato — backend mock

Non ospitato su LLM EU in questa distribuzione: le richieste sono risposte dal backend mock integrato nella regione dev-mock. Nessun peso viene eseguito.

Riepilogo

Whisper large-v3 is OpenAI’s multilingual speech-recognition and speech-translation model with 1.54B parameters under the Apache-2.0 licence. It has a fixed 30-second receptive field and processes longer audio with a sliding window, and it uses 128 mel frequency bins rather than the 80 of earlier versions. The model card warns that it was evaluated on roughly ten languages and should be evaluated in-domain before deployment.

Specifiche

Fornitore
OpenAI (US)
Finestra di contesto
Non verificato: il fornitore non ha documentato questa affermazione.
Parametri
1.54B
Modalità
audio
Licenza
apache-2.0 · licence
Pesi aperti
yes
Data di rilascio
07 nov 2023
Attività
translation, multilingual-support, summarization

Lingue

Forte: en, de, fr, es, it, pt, nl, pl

Adeguato: sv, da, fi, cs, ro, el, uk, ru, tr, ar, zh, ja, ko, hi

Lingue

Forte

  • en
  • de
  • fr
  • es
  • it
  • pt
  • nl
  • pl

Adeguato

  • sv
  • da
  • fi
  • cs
  • ro
  • el
  • uk
  • ru
  • tr
  • ar
  • zh
  • ja
  • ko
  • hi

Non verificato: il fornitore non ha documentato questa affermazione. — a language is listed as strong only where a source claims it; the card links the evidence where one exists.

Quando usarlo

Punti di forza

  • one model for transcription and translation to English
  • long-standing tooling support across every major inference stack
  • Apache-2.0 licence

Limiti

  • fixed 30-second receptive field; longer audio needs chunking logic
  • the card only claims strong recognition for about ten languages
  • the card advises against using it for classification or speaker attributes