Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele
Idanwo Idanwo Ọrọ whisper
Review Whisper’s multilingual audio-to-text design, its token-selected tasks, and what broad training coverage does and does not guarantee.
Ibeere 1 ti 8
O fẹrẹ to wakati melo ti ohun ohun ti a ti kọ Whisper lori?
Awotẹlẹ gbogbo ibeere8 ibeere
Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.
- O fẹrẹ to wakati melo ti ohun ohun ti a ti kọ Whisper lori?
- 680,000 wakati
- 6,800 wakati
- 68,000 wakati
- 6.8 milionu wakati
- Aṣoju agbedemeji wo ni Whisper ṣe iṣiro lati inu ohun aise ṣaaju ki kooduopo naa?
- Aise igbi tensor
- A log-Mel spectrogram
- Awọn ẹya delta MCCC nikan
- Fekito foonu-gbona kan
- Bawo ni awoṣe Whisper kan ṣe ṣe awọn iṣẹ-ṣiṣe pupọ bi transcription ati itumọ?
- O fifuye awoṣe ti o yatọ fun iṣẹ-ṣiṣe kan
- O nṣiṣẹ lọtọ itanran-aifwy olori ni afiwe
- Awọn ami itọka pataki pato iṣẹ-ṣiṣe lakoko iyipada
- Awọn olumulo satunkọ spectrogram pẹlu ọwọ
- Kini faaji nkankikan gbogbogbo wo ni Whisper lo?
- Nẹtiwọọki ti ariyanjiyan-nikan
- Akopọ LSTM loorekoore
- A awoṣe itankale
- Ayipada kooduopo-decoder
- Kini idi ti Whisper ni pataki paapaa logan ni akawe si awọn eto ASR iṣaaju?
- O ti gba ikẹkọ lori oniruuru, ohun afetigbọ oju opo wẹẹbu alariwo
- O ṣiṣẹ nikan lori ohun afetigbọ-situdio
- Ko lo ikẹkọ ẹrọ
- O nilo gbohungbohun aṣa
- Labẹ iwe-aṣẹ wo ni a ti tu awọn iwuwo awoṣe Whisper silẹ?
- Ohun-ini, API-nikan
- MIT (orisun ṣiṣi)
- GPL laisi lilo iṣowo
- Creative Commons ti kii-itọsẹ
- Approximately how long is each audio segment Whisper converts to a log-Mel spectrogram?
- 3 seconds
- 3 minutes
- 30 seconds
- 30 minutes
- What does Whisper’s English-translation task produce from non-English speech?
- A translated recording in the speaker’s voice
- Only a transcript in the original language
- A list of detected speakers
- An English text translation