AudioLDM 2
Free.ai (self-hosted)
·
music
·
~5000 Ama-token ngalinye track
·
3.9
kusuka 8 Abasebenzisi balolu hlu
AudioLDM 2 yi an imodeli yokukhiqiza umculo eyenziwe ngu Haohe Liu. Onamandla kakhulu ku-Music generation from text. I-self-hosted ku-Free.ai GPUs - isebenza ngokukhululekile ngaphezu kwe-token pool yakho yansuku zonke (5,000 tokens Ingoma ngayinye). Kukhishwa ngaphansi kwe-Apache 2.0 - ukusetshenziswa kokuthengiswayo kuvunyelwe ku-Free.ai.
Sebenzisa nge-API
I-REST API ehambisana ne-OpenAI. Dala isithonjana bese ubiza le modeli emaminithini.
curl -X POST https://api.free.ai/v1/music/generate/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"audioldm2","prompt":"your prompt here"}'Amamodeli afana
Imibuzo ebuzwa kaningi
AudioLDM 2 ikhiqiza umculo omusha wezinhlamvu (nakwezinye izinjini, izingoma) kusuka kumbhalo osheshayo. Yazisa imizwa, i-genre, ukudweba, noma i-BPM futhi AudioLDM 2 ibuyisela umsindo oqediwe.
10 kuya ku-60 imizuzwana ngayinye yohlobo ngokuya ngenjini. Ukwenza amathuluzi ade, hlanganisa izihlobo eziningi nge /music/stitch/ noma yenza iziqephu ezingu-30 zesikondi bese uqala ukufayela.
Ezinye izinjini zihlanganisa izingoma ezivela kumazwi akho (faka amazwi efomu); ezinye ziyizinhlamvu kuphela. Khangela ifomu kulekhasi - indawo yemazwi isho ukuthi izingoma zixhaswe.
Pop, rock, EDM, jazz, ambient, classical, lo-fi, hip-hop instrumentals zonke zibonisa ngokunengqondo. Izinhlobo ze-niche (gamelan, Carnatic, ukudansa kwethonya) zincane - zichaza amathuluzi ngokucacile ukuze zithole imiphumela engcono.
AudioLDM 2 ihamba ku-GPU yethu yangempela ephikisana ne-pool yakho yamahhala yansuku zonke kuqala; $5 → 200,000 i-token ekhokhelwa ngemuva kwalokho. Ngokungafani ~5,000 i-token ngayinye.
WAV ngokuzenzakalela (akukho phutha, ilungile ukuxhuma okuningi). MP3 itholakala ku-output picker yefayela elincane.
No. AudioLDM 2 ibuyisela umculo oxhumene. Ukuwuhlukanisa ube izimbali, sebenzisa isingeniso /music/stem-splitter/, esikhipha izingoma, ama-drums, ama-bass nezinye.
Yebo - /batch/ ithatha uhlu lwezicelo. Ingoma ngayinye iwela ku /account/?tab=history nesicelo njengegama lefayela. Ukulawula okuningi, sebenzisa i-API.
Yebo - POST isinqumo sakho ku /v1/music/generate/ ngemodeli "AudioLDM 2". /api/ inesimo esigcwele esifaka phakathi i-lyric, isikhathi, ne-tempo parameters.
AudioLDM 2 ayikhiqizi ama-copyrighted compositions - iziphume zikhiqizwa ngokutsha. Free.ai inikeza ukusetshenziswa okuhweba; ugcina ilungelo lemisindo yakho ekhiqizwe. Ukubiza umculi kwi-prompt buza "kwi-style ye" - umsindo okhona uvela ekuqaleni.
15 imizuzwana kuya kumizuzu engu-2 ngayinye yengoma ngokuya ngesikhathi + injini. Amamodeli epremium aqedela ngokushesha. Sebenzisa inkinobho yokulinganisa ukuvala i-tab - imiphumela iwela ku /account/?tab=history uma kwenziwe.
Yebo - iphutha libuyisela emuva ngokuzenzakalela. Ama-tracks ahamba kodwa azwakala anzima, thumela i-imeyili contact@free.ai nge-link yokusibelana futhi sizobhekana nesimo ngasinye.