Maswali yanayounganishwa na mwongozo · Kati Kiwango

OpenAI Maswali ya kunong'ona

Kagua muundo wa data ya mafunzo ya Whisper, tabia ya lugha nyingi, na hali za kutofaulu zilizorekodiwa bila kudhani kuwa nakala ni sahihi kila wakati.

Njia za mwongozo zinazohusianaKunong'ona kwa Openai
Swali 1 ya 8

Je, lengo kuu la OpenAI ni nini?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Je, lengo kuu la OpenAI ni nini?
    • Utambuzi otomatiki wa usemi: kunakili na kutafsiri sauti
    • Inazalisha picha kutoka kwa maandishi
    • Kuandika nambari ya kompyuta
    • Kuunda uhuishaji wa 3D
  2. What was true of the language balance in Whisper’s training audio?
    • Every language contributed the same number of hours
    • The dataset was English-heavy, with much less audio for most languages
    • The dataset contained only English audio
    • Most data was studio-recorded speech in one language
  3. What does Whisper use to condition the decoder on the requested task and language?
    • A separate camera feed
    • A hand-written transcript of the answer
    • Special tokens in the decoder prompt
    • A unique model checkpoint for every utterance
  4. Ni uwezo gani ambao Whisper NOT asili hutoa?
    • Kutafsiri hotuba kutoka lugha nyingine hadi Kiingereza
    • Inatambua ni lugha gani inayozungumzwa
    • Kuongeza mihuri ya muda kwa manukuu
    • Inazalisha video ya kweli kutoka kwa kidokezo
  5. Which long-form transcription problems does OpenAI’s paper describe?
    • Repetition loops and missing words near segment edges
    • Perfect diarization of every speaker
    • Automatic correction of every proper noun
    • Guaranteed removal of background music
  6. Je, ni udhaifu gani unaojulikana wa Whisper kutazama?
    • Haiwezi kuchakata sauti yoyote ya Kiingereza
    • Mara kwa mara inaweza 'hallucinate' maandishi ambayo hayakusemwa
    • Inatoa sauti tu, sio maandishi
    • Inakataa kuongeza alama za uakifishi
  7. Approximately what share of Whisper’s training data was English audio paired with English transcripts?
    • About 18%
    • Takriban 1%
    • About 65%
    • About 100%
  8. What use does OpenAI caution against when transcription errors could have serious consequences?
    • Noncommercial experimentation with public samples
    • Kuongeza mihuri ya muda kwa manukuu
    • Using the model to identify a language
    • High-risk decision-making without appropriate verification