Dia Voice Clone
Free.ai
·
voice
·
~500 tohu i ia clip
Ko Dia Voice Clone ko a Tauira AI. I whakateretia mā ngā tauira o waho — ~500 ngā tohu i ia whakamahinga (50% te tohu i runga i te utu o te awa).
Ka whakamahia mā te API
API REST OpenAI-hōia. Whakana tētahi kī, ā, ka karangatia tēnei tauira i roto i ngā takirua.
curl -X POST https://api.free.ai/v1/voice/clone/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"premium/dia-tts/voice-clone","text":"hello world","audio_url":"https://..."}'E pā ana ngā pātai
E tautoko ana a Dia Voice Clone i tētahi awhe reo whānui. E whakawhirinaki ana te rārangi tika ki te mīhini; ka whakaaetia e te tāurunga i tēnei pātaka ētahi kupu, ā, ka whakawātea te mīhini i roto i ōna reo tautoko. Tirohia te /rohe/ mō te kōwhiringa pūkaha maha katoa mēnā e hiahiatia ana e koe tētahi reo tauwhāiti.
Ko te nuinga o ngā mīhini e whakawātea ana i te reo Ingarihi ā-American-pūmau mā te pūnaha, ā, he āhuahira ā-rohe e tika ana mō ngā reo kāore i te Ingarihi.
He rerekē te tautoko a SSML i runga i te mīhini. Ko ngā tohu whakaroa, ngā tohu whakaroa, me ngā tohu whakahua i honoa ki te nuinga o ngā mīhini utu, ki ētahi o ngā mea whakanohoia. Ka mahi tonu te kupu pūnoa — kāore he tohu e hiahiatia ana.
Tērā te TTS Streaming i runga i ngā mīhini utu mā te /v1/tts/ API te wāhi mutunga me te rerenga = pono. Ka hoki te UI Wīwī i tēnei pātaka ki te tātari katoa ina oti te whakarārangi.
He mīhini TTS utu nui a Dia Voice Clone. Ko te utu e whakarea ana me te tatau āhua — i te nuinga o te wā ~30 ngā tohu mō ia āhua. $1 e hoko ana i ngā tohu 750,000, nō reira he $5 te paki e taupoki ana i ngā rau tau o ngā āhua.
Tae atu ki te 5,000 ngā tohu i ia tono i runga i te UI Wīwī. Mō ngā wāhanga roa ake (ngā pukapuka oro, ngā wāhanga katoa), ka whakamahia te /rohe/rangirangi/ e whakawātea ai me te whakawātea i ngā wāhanga, ka karanga rānei te API i roto i tētahi porowhita.
He — POST tētahi rārangi aho ki te /v1/tts/batch/, te whakamahi rānei i te UI mokowā mahi i te /workspace/ hei whakarārangi i te TTS ki roto i tētahi pūtautau roa ake (hei tauira, whakamāori → kōrero → whakarārangi).
He — POST kupu ki te /v1/tts/ me te tauira "Dia Voice Clone" (te slug rānei i tēnei pātaka). Ka hoki ki WAV, MP3 rānei. Tirohia te /api/ mō te tohutoro katoa + ngā tātai SDK.
He kupu-ki-te-kōrero tēnei pātū, ehara i te tārua reo — ko te reo te pūnaha pūnaha. Mō te tārua reo (te whakataki i tētahi oro tohutoro), tirohia te /rongo/kōrero/, e hiahiatia ana kia whai mana ā-rongo koe, kia whai whakaaetanga tuhituhi rānei.
Ko ngā mīhini whakanoho-i-te-pono e haere ana i runga i ngā GPU Free.ai-whāiti; kāore he mea e mahue ana i a tātau māngai. Ko ngā mīhini Premium e whakawhiti kōrero ana ki ngā kaiāwhina tauira ki runga i a tātau DPA. Kāore e whakaakona e tātau i ōna tāurunga, ā, kāore e hokona i te raraunga.
He — Free.ai e whakaae ana ki te whakamahi hokohoko o te oro i hangaia. Ko te whakaaetanga o te mīhini taketake (Apache 2.0, MIT, ngā whakaritenga kaihokohoko rānei) e whakaaturia ana i runga ake nei, i runga anō i te pātū tohutoro tauira; i roto i te mahi, ko te tikanga o tēnei he kōrero, he whakapuakitanga, he podcast, me ngā taupānga katoa i roto i te tauine.
He — he whakawhiwhinga ā-pūnaha i ngā mahi whakapeka ki te pūtake (pūnaetanga i ia rā, tohu whakawhiwhi rānei). Mēnā kāore he whakawhiwhinga i te rā kotahi, tātai mai ki te tātai@free.ai.