CosyVoice 2
Free.ai (self-hosted)
·
tts
·
~500 tokens per clip
·
4.3
daga 3 users of this category
CosyVoice 2 shi ne a Voice-to-speech wanda aka gina da Alibaba FunAudioLLM. Low-latency streaming TTS, cross-lingual voice cloning, real-time voice.. Self-hosted a kan Free.ai GPUs - runs free a kan ku yau da kullum token pool (500 tokens per clip). An saki a karkashin Apache 2.0 — amfanin kasuwancin da aka yarda a kan Free.ai.
Use via API
curl -X POST https://api.free.ai/v1/tts/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"cosyvoice2","text":"hello world"}'Tambayoyi da ake yi da yawa
CosyVoice 2 yana goyon bayan yarukan da dama. Wannan jerin ya danganta da injin; fuskar wannan shafi tana karɓar duk wata rubutu kuma injin zai yi zane cikin yarukan da yake goyon baya. Ka duba /voice/ domin zaɓaɓɓen injin-da-da yawa idan kana buƙatar wata harshe.
Most engines render neutral-American English by default and a region-appropriate accent for non-English languages. Premium engines may expose accent variants — paste a sample to compare.
SSML support varies by engine. Pause, prosody, and emphasis tags are honored on most premium engines and on a few self-hosted ones. Plain text always works — no markup required.
Ana iya amfani da TTS mai gudana a kan injiniyoyin premium ta hanyar /v1/tts/ API na ƙarshen maɓalli tare da stream=true. Haɗin gwiwar mai amfani da yanar gizo a wannan shafi yana mayar da duk clip bayan an kammala nunawa.
CosyVoice 2 yana tafiya a kan GPUs ɗinmu. Jinsi yana rufe daga pool ɗinka na yau da kullum na kyauta a farko. Bayan an cika, alamun biyan kuɗi suna farawa a $5 → 200,000 alamun kuɗi. Kashi ~5 alamun kuɗi ga kowane haske, mafi ƙarancin 100 ga kowane bidiyo.
Har zuwa 5,000 na haruffa a kowace tambaya a kan haɗin gwiwar mai amfani da yanar gizo. Ga sassa masu tsawo (litattafai masu sauti, dukkan sassa), amfani da /voice/audiobook/ wanda ke yin kwashewa da haɗewa da kai tsaye, ko kira API cikin wani layin.
Ya - POST wani jerin waƙoƙi zuwa /v1/tts/batch/, ko amfani da filin aiki UI a /workspace/ don haɗa TTS cikin wani tsawo na ruwa (misali, fassara → magana → stitch).
Na'am — POST rubutu zuwa /v1/tts/ da model="CosyVoice 2" (ko slug a wannan shafi). Yana mayar da WAV ko MP3. Ka duba /api/ don cikakken bayani + SDK snippets.
Wannan shafi na rubutu zuwa magana ne, ba na kwaikwayon magana ba — magana ita ce diff ɗin injin. Don kwaikwayon magana (ta hanyar shigar da wani sauti na alaƙa), ka duba /waƙa/kwaikwayo/, wanda ya buƙaci ka mallaki haƙƙin magana ko kuma ka sami izinin rubutacce.
Self-hosted engines run on Free.ai-owned GPUs; nothing leaves our servers. Premium engines pass text to upstream model providers under our DPA. We do not train on your inputs and do not sell data.
Na'am — Free.ai na ba da izinin amfanin kasuwanci na sauti da aka samar. An nuna lasisi na injiniya (Apache 2.0, MIT, ko kuma sharuddan mai sayarwa) a sama da kuma a kan shafi na alamar misali; a cikin aiki wannan yana nufin waƙoƙin, tallace-tallace, podcasts, da kuma aikace-aikace suna cikin fagen.
Na'am — aiki da ya kasa yin aiki da shi zai sake dawo dashi zuwa wurin da ya samo shi (ta hanyar ajiyar yau da kullum ko kuma ta hanyar biyan kuɗin. Idan an sake dawo da shi ba zai nuna ba a ranar da ta gabata, aika imel zuwa contact@free.ai.