Wizper (Whisper v3)
Free.ai
·
stt
·
~500 tokens per minute
Wizper (Whisper v3) ke a Mofuta oa puo ho ea ho mongolo. Routed through external models - ~500 tokens ka metsotsoana (50% markup over upstream cost).
Ho sebelisa ka API
REST API e lumellanang le OpenAI. E etsa konopo'me e bitsa mofuta ona ka metsotsoana.
curl -X POST https://api.free.ai/v1/stt/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"premium/wizper","audio_url":"https://..."}'Lipotso tse botsoang khafetsa
Wizper (Whisper v3) e ngola molumo o boleloang ho ea ho mongolo. Kopitsa MP3, WAV, M4A, kapa faele ea video'me Wizper (Whisper v3) e khutlisa moelelo o feletseng le lihlooho tsa SRT / VTT tse sa tšoaneng le timestamp.
Wizper (Whisper v3) handles dozens of languages - Whisper-family models cover 90+, others vary. Pick "auto-detect" or specify the language for highest accuracy.
The rate of word-error is 5-10% on clean English audio, 10-20% on noisy or accented audio. Liphetoho tse kholo tsa ts'ebetso e ts'oanang li etsa hore li be betere ka liketsahalo tse thata - khetha tse kholo ha li-audio li le thata.
Ee - karolo e ngoe le e ngoe e na le li-timestamps tsa ho qala/ho fela. E-ba le SRT kapa VTT'me li-timemap li tla hlahisoa ka ho toba ho video ea hau.
Wizper (Whisper v3) ke mochini oa ho ngola oa premium. Ka li-tokens tsa ~ 500-1,500 ka metsotsoana ea audio, tse bonts'itsoeng ka nako e le 'ngoe pele o etsa. Ho ngola ka bomong ho sebetsa ka ntle ho 30,000-token ea hau ea letsatsi le letsatsi; li-models tsa premium li lefa-ka-ho-e-na le-go, le top-ups ho tloha $ 1.
MP3, WAV, M4A, FLAC, OGG, hammoho le video (MP4, MOV, WebM) - re tlosa audio. Max 500 MB ka ho kenya. Lifaele tse telele? Tsebisoa ka /audio/cut/ kapa u sebelise /v1/stt/batch/.
Li-diaries tsa motsamaisi ke litsela tse fapaneng — toggle "diarize" ho /transcribe/. Wizper (Whisper v3) e sebetsana le ho ngola; li-diaries li tšoaea li-segments ka ho toba le Motsamaisi 1 / Motsamaisi 2 / jj.
Ee - /batch/ e amohela lebokose la lifaele tsa molumo. Lingoloa tsohle li tla ba /account/?tab=history le lebitso la faele la mantlha. Ho boloka moqhaka oa lebokose la lifaele, sebelisa API.
E-na - POST audio ea hau ho /v1/stt/transcribe/ le model="Wizper (Whisper v3)". E khutlisa JSON le mongolo + li-segments + timestamp tsa boemo ba mantsoe. /api/ e na le ho ngolisoa ka ho feletseng.
Li-model tse sebete li boloka molumo ho GPU ea rona; premium e tsamaea ka DPA. Molamu o tlosoa ka mor'a lebokose la ho arolelana (24h anon, 7d login). Re sa koetlise ka ho kenya ha hao.
Ee - Free.ai e fana ka ts'ebeliso ea khoebo ea li-transcripts. U hloka litokelo tsa audio eo u e tlisitseng (ho ngolisoa ha hau, thepa e lekanyelitsoeng, kapa litaba ka tumello).
Likarolo tsa nako ea 'nete ke ka bongata ba 0.05-0.2× - podcast ea metsotso e 60 e ngola metsotso e 3-12. Li-models tsa Premium li ka qetella li potlakile. Sebelisa konopo ea ho khutlela pele ho koala tab.