Distil-Whisper large-v3
Free.ai (self-hosted)
·
stt
·
~500 Whakahauhau minute
Ko te Distil-Whisper large-v3 he a Tauira kōrero ki te tuhituhi i hangaia e te HuggingFace. He kaha rawa i te Real-time transcription, large-volume batch STT.. I whakanohoia ki ngā GPUs Free.ai - e haere ana i te wātea i runga i tō tātau pūpū tohu i ia rā (500 ngā tohu i ia wa). I tukua i raro i te MIT - ka whakaaetia te whakamahinga hokohoko i runga i te Free.ai.
Ka whakamahia mā te API
API OpenAI-hōu REST. Whakanao i tētahi kī, ā, ka karangatia tēnei tauira i roto i ngā waeine.
curl -X POST https://api.free.ai/v1/stt/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"distil-whisper-large-v3","audio_url":"https://..."}'He ōrite nga tauira
He pātai pānui
Ka whakamārō te Distil-Whisper large-v3 i ngā oro kōrero ki te kupu. Whakarewa i tētahi MP3, WAV, M4A, faila ataata rānei, ā, ka hoki te Distil-Whisper large-v3 ki te whakamārō katoa, tae atu ki ngā whakarārangi SRT/VTT kōwhiria me ngā tātaitai wā.
Distil-Whisper large-v3 handles dozens of languages - Whisper-family models cover 90+, others vary. Pick "auto-detect" or specify the language for highest accuracy.
Ko te ōrautanga hē-wānanga-wānanga ko te 5-10% i runga i te oro Ingarihi mārō, 10-20% i runga i te oro whakahauhau, i te oro whakahauhau rānei. He nui ake ngā rerekētanga o te hanganga ōrite i ngā take uaua - ko te kōwhiringa nui ake ina pōturi te oro.
He - kei roto i ia wāhanga ngā tātai tīmata/whakamutu. Ka whakawātea hei SRT, VTT rānei me te mahere wā ki runga i tōmu ataata.
Ka haere a Distil-Whisper large-v3 i runga i a tātau ake GPUs ki tōtou pū waikawa wātea i te tuatahi; $5 → 200,000 ngā tohu whakawhiwhita i muri mai. Tata ki ngā tohu ~500 i ia minu.
MP3, WAV, M4A, FLAC, OGG, tae atu ki te ataata (MP4, MOV, WebM) - ka tangohia e tātau te oro. Te nui rawa o te 500 MB i ia tāpiritanga. He nui ake ngā faila? Whakawehe me / oro/ whakaiti /, hoatu rānei /v1/stt/batch/.
He whakawhitinga motuhake te whakahua i te kaikōrero — ka whakawhiti "pāpāho" i runga i te / whakamāori /. Distil-Whisper large-v3 e whakahaere ana i te whakamāoritanga; ko te whakahua i ngā tohu i ia wāhanga me te Kaikōrero 1 / Kaikōrero 2 / ērā atu mea.
He - /batch/ e whakaae ana ki tētahi pūrākau o ngā pūranga oro. Ka tae ia whakahua i roto i /account/?tab=hitori me te ingoa pūranga taketake. Mō te tiaki i te rākau pūrākau, ka whakamahia te API.
He - POST tō tou oro ki te /v1/stt/transcribe/ me te tauira "Distil-Whisper large-v3". Ka hoki mai te JSON me te kupu + wāhanga + wā-wā. He tohutoro katoa te /api/.
Ka pupuri ngā tauira ā-iwi i te oro i runga i a tātau GPU; ka haere te utu mā te DPA. Ka whakakoretia te oro i muri i te matapihi tiritiri (24h anon, 7d te whakakite-i). Kāore e whakaakona e tātau i ōna tāuru.
He - ka whakaaetia e te Free.ai te whakamahinga hokohoko o ngā tātai. E hiahiatia ana e koe ngā mana ki te oropū i whakarewaina e koe (tātou whakataki, rawa whakawhiwhia, ihirangi rānei me te whakaaetanga).
Ko te take o te wā tūturu tata tonu ki te 0.05-0.2× - he podcast 60-meneti e whakarerekē ana i roto i te 3-12 minute. He tere ake te mutunga o ngā tauira Premium. Ka whakamahia te kione whakarerekē hei kati i te tīpako.