AudioLDM 2
Free.ai (self-hosted)
·
music
·
~5000 Whakahauhau track
·
3.9
mai i 8 Ka taea te whakawātea i te taupānga o te kāwai.
Ko te AudioLDM 2 he a Tauira whakawhanake pūoro i hangaia e te Haohe Liu. He kaha rawa i te Music generation from text. I whakanohoia ki ngā GPUs Free.ai - e haere ana i te wātea i runga i tō tātau pūpū tohu i ia rā (5,000 ngā tohu i ia whakataki). I tukua i raro i te Apache 2.0 - 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/music/generate/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"audioldm2","prompt":"your prompt here"}'He ōrite nga tauira
He pātai pānui
Ka whakaputaina e te AudioLDM 2 te pūoro pūoro taketake (me ētahi mīhini, ngā pūoro) mai i tētahi whakahau tuhituhi. E whakaahua ana i te āhua, te momo, te pūoro, te BPM rānei, ā, ka hoki mai te AudioLDM 2 ki tētahi whakahau oro.
10 to 60 seconds per generation depending on the engine. For longer pieces, chain multiple generations through /music/stitch/ or generate in 30-second segments and crossfade.
Ko ētahi mīhini e whakakotahi ana i ngā pūoro mai i ngā kuputuhi e tuku ana koe (whakapae ngā kuputuhi i roto i te āhua); ko ētahi he pūoro anake. Tirohia te āhua i tēnei pātaka - ko te tikanga o te āpure Lyrics ko te tautoko i ngā kuputuhi.
Ko ngā pūoro Pop, Rock, EDM, Jazz, Ambient, Classical, Lo-Fi, Hip-hop katoa e tino mārama ana, ā, ko ngā momo āhuahira (gamelan, Carnatic, te waiata, te waiata me te whakatangitangi) he iti iho - e whakaahua tika ana i ngā pūoro mō ngā hua pai rawa.
Ka haere a AudioLDM 2 ki a tātau ake GPUs ki tōna pūngao wātea i te tuatahi; $5 → 200,000 ngā tohu whakawhiwhita i muri mai. Tata ki te ~5,000 ngā tohu i ia whakarongo.
WAV mā te pūmau (kore ngaro, e whakaritea ana mō te whakaminenga anō). Ka wātea te MP3 i roto i te kōwhiringa huaputa mō ngā pūkete iti iho.
Ko te kore. AudioLDM 2 e hoki ana ki tētahi whakarerekētanga. Hei wehe i te pūtake, ka haere te huaputa mā /pūoro/pūtake-whakahaere/, e whakawātea ana i ngā kōrero, ngā pūrere, te pū me ētahi atu.
He - /batch/ e whakaae ana ki tētahi rārangi whakahauhau. Ka tae ia whakahauhau ki /account/?tab=hitori me te whakahauhau hei ingoa faila. Mō ētahi atu whakahaere, ka whakamahia te API.
He - POST tō tātou whakahau ki te /v1/music/generate/ me te tauira "AudioLDM 2". He tohutoro katoa te /api/ tae atu ki ngā kupu, te roanga, me ngā tātai whakatere.
Kāore e whakawātea ana a AudioLDM 2 i ngā hanganga whakawhanake - ka whakawāteatia ngā huaputa. Ka whakaaetia e te Free.ai te whakamahinga hokohoko; e pupuri ana koe i ngā mana ki ngā aho i hangaia. Ko te whakaingoatanga o tētahi kaitapere i te whakahau e pā ana ki "i roto i te kāhua o" - ko te orotanga tūturu te taketake.
15 ngā taki ki te 2 ngā minu mō ia whakataki, i runga anō i te roanga + mīhini. Ka tere ake te mutunga o ngā tauira Premium. Ka whakamahia te kīanga kōaro hei kati i te tīpako - ka noho te hua ki /account/?tab=hitori ina oti.
Yes - failed renders auto-refund. For tracks that ran but sound bad, email contact@free.ai with the share link and we will review case-by-case.