VibeVoice

Speaker 4

Paerewa English Neutral VibeVoice

He oro AI a Speaker 4 i whakahaua e te tauira kupu-ki-whakaahua VibeVoice. E kōrero ana tēnei oro tau tōmua i te English, ā, ka tuku i te whakakotahitanga reo Whare whetū-quality. Me te tere whakawhanake tata tonu me te arotakenga āhuatanga o 5/5, he pai te Speaker 4 mō podcasts, dialogues, long-form narration, multi-speaker content. Kua hangaia te VibeVoice me te Microsoft under the MIT license, e haumaru ai mo te whakamahinga hokohoko. Kei roto i ngā kaha kī: {whakamahi}.

Kāore he arotakenga

VibeVoiceWhakamāramatanga tauira

Kāhua VibeVoice
kaiwhakawhanake Microsoft
Whakahautanga
Āhuatanga Tūturu
Whakawhiwhinga MIT
Ko te tārua Kāore i te wātea
Te āhua Paerewa (2 pūtea/1K pūāhua)
Parameter 1.5B
Hanganga LLM + DAC
Ko te raraunga whakaakoranga 100000 wae
Taone 2025

Ko ngā take whakamahi tino pai mō Speaker 4

Ko ngā taupānga i whakaritea i runga i tēnei reo

He pukapuka oro me te kōrero

Ka whakamahia e Speaker 4 hei kōrerorero i ngā ihirangi āhua roa me te pūāhua me te kīanga māori.

Whakapāpāpānga o te wikiō

E tāpiri ana i te kōrero mātauranga ki ngā pouaka whakaata YouTube, ki ngā pāpāho, me ngā ihirangi pāpāho pāpori.

Āhuatanga o ngā taupānga

Mā te whakatūnga tere e tino pai ai tēnei reo mō ngā taupānga wā tūturu, ngā kaiako mata, me ngā utauta āhei ki te uru.

Podcasts & Whakaputanga

He pai te huaputa o te āhua o te whare taupuni mō ngā podcast, ngā irirangi, me te whakapāpāho ngaio.

He nui ake VibeVoice Pāpāho

Ko ētahi atu reo mai i te tauira TTS ōrite

Speaker 1

English Neutral

Speaker 1 (Chinese)

Hainamana Neutral

Speaker 2

English Neutral

Speaker 2 (Chinese)

Hainamana Neutral

Speaker 3

English Neutral

E pā ana ngā pātai

E rua ngā momo VibeVoice a Microsoft: he tauira 1.5B mō ngā ihirangi ā-rohe roa (tata ki te 90 min, 4 ngā kaikōrero) me tētahi tauira 0.5B o te wā tūturu mō te whakawhitinga me te ātete reo tuatahi ~200ms. Ko te momo 1.5B e tino pai ana i ngā podcast me ngā pukapuka oro me te ōritetanga o te kaikōrero i ngā wā roa. Whakama: I tangohia e Microsoft te waehere TTS mai i te puna, ā, ko te oro i hangaia, ko ngā whakawāteatanga AI e taea te whakarongo.

I hangaia a VibeVoice e Microsoft, ā, ka tukua i raro i te whakaaetanga a MIT (whakahauhau-whakahauhau anake), e whakaae ana ki te whakamahi hokohoko o te oro i hangaia.

E tautoko ana a VibeVoice i te reo 1: English.

Ko VibeVoice i roto i te taumata Whakawhiwhinga — 4 ngā pūtea i ia 1,000 ngā āhuatanga. Ka taea e koe te tohutoro i tētahi reo VibeVoice mō te wātea i mua i te whakanaotanga o te oro katoa.

He tere whakawhanake ā-papa te VibeVoice. He maha nga wa e hiahiatia ana te whakawhanaketanga, i runga anō i te roanga o te kupu.

Kua whakawāteatia a VibeVoice ki te 5/5 mō te āhuatanga oro i runga i te TTS.ai. Ka tuku i te whakaakoranga, te kōrero ā- tangata.

Kāore, ka whakamahia e VibeVoice tētahi huinga pūoro whakatū. Mō te tārua reo, ka whakamātau i ngā tauira pēnei i te CosyVoice 2, GPT-SoVITS, Chatterbox rānei.

Heoi, kua whakapuakina te VibeVoice mō ngā podcasts, ngā pukapuka oro, ngā ihirangi kōrero maha-whakahaere. Ko ōna kōrero maha, tae atu ki te 90 min, ko ngā kaha whakawhanake podcast e noho ana he kōwhiringa tino pai mō tēnei take whakamahi.

Heoi, kua whakaaetia a VibeVoice i raro i te MIT (whakahauhau-whakahauhau anake), e whakaae ana ki te whakamahi hokohoko. Ka taea te whakamahi i te oro i hangaia ki ngā reo VibeVoice i roto i ngā ataata, podcasts, taupānga, kēmu, me ētahi atu kaupapa hokohoko.

Heoi, ko ngā oro katoa i runga i te TTS.ai e whakamahi ana i ngā tauira pūtake tūwhera-pūnaha-pūnaha (MIT, Apache 2.0). Ko te oro i hangaia ko tātou hei whakamahi i roto i ngā ataata, podcasts, ngā taupānga, ngā kēmu, me ētahi atu taupānga hokohoko.

Whakahaua tētahi tono POST ki te /api/v1/tts/ me te ingoa tauira me te tohu reo. Tirohia tātau pou Papatono API mō ngā tauira waehere i roto i te Python, JavaScript, Go, me te cURL.

Heoi, ka kōwhiria te kī taupānga i tēnei pātaka hei whakarongo i tētahi tauira. Ka taea hoki e koe te tuhituhi i te kupu ā-ringa i runga i te pātaka Whakawhitiwhiti me te whakaputa i tētahi tirohanga wātea me ētahi reo.

Whakamātautau Speaker 4 Ināianei

Type i tētahi kupu me te mōhio ki a ia e kōrero ana Speaker 4. Waihoki ki te whakamahi.