E mōhio ana ngā tohu ki te tauira i kōwhiria - ka kōwhiria kia whakawātea tētahi ki roto i tōna kupu i reira ka puta ai:
Ka pānui tēnei tauira i te kupu noa, nā reira ka whakakāhoretia ngā tohu ā-waitara. Mō te āhua o te tohu-taihi, ka huri ki tētahi tauira whakamārama pēnei i a Orpheus, Bark rānei.
Ko ngā tohu whakawhiwhinga reo māori. I whakawhiwhia e Qwen3-TTS i tēnei rā; Ka tae mai ētahi atu tauira whakamāramatanga.
Kāore tēnei tauira i te tautoko i ngā tohutohu kāhua — ka huri ki tētahi tauira e mahi ana (hei tauira, Qwen3-TTS).
Ka tautuhia ngā tohutohu ā-ringa (wāhi = tohutohu):
-12
+12
Ka whakahaeretia te whakahuahua (0 = pūmau, 1 = tino whakahuahua)
Te taumahatanga tohu whakawātea-whakahaere (nui ake = nui ake te whakahau-whakahaere)
E whakaahua ana i te kāhua reo i roto i te reo māori (ka whakamahia e te Pāpāho tēnei i te wāhi o ngā reo i whakaritea i mua)
He pūāhua taupānga te taupānga taupānga: Ka whakamahia ngā tohu [S1] me [S2] hei tohu i ngā kaikōrero rerekē. Hei tauira: [S1] Hei hei! [S2] Hei, he pēhea koe?
Darwin-TTS-1.7B-Cross by FINAL-Bench is a research variant of Qwen3-TTS-1.7B with an unusual construction: 84 of its talker-FFN tensors (about 8.6% of them) are blended at a 3% ratio with the matching tensors from Qwen3-1.7B-Base, all without any retraining. The result is a model that produces noticeably crisper cross-lingual voice cloning across Korean, English, Japanese, and Chinese — its four core languages. It operates in zero-shot voice-clone mode, needing only about three seconds of reference audio to capture a speaker. Darwin is best suited to transferring a single reference voice across those four languages, for example dubbing or multilingual narration with consistent speaker identity.
Pai mo: Cross-lingual voice cloning between English / Korean / Japanese / Chinese with a single reference voice
Darwin starts from Qwen3-TTS-1.7B but blends a small fraction of its talker-FFN weights with the matching weights from the Qwen3-1.7B base language model. This training-free blend sharpens cross-lingual voice cloning rather than changing the base voices.
English, Korean, Japanese, and Chinese. The FINAL-Bench release specifically markets its cross-lingual blend for those four, and the deployed model ships voices for them.
About three seconds. It works in zero-shot mode, so no fine-tuning or training is required — you provide a short reference clip and it generates new speech in that voice.
E tāuru ana i tōna mairā hei haere tonu i te whakaputa kōrero mō te wāteatanga, hei whakaingoatanga rānei mō ngā pūāhua tāpiri 15,000.
Kāore he kupu whakarongo e hiahiatia ana — ka tukuna mai e tātau he pātahitanga ki te tautuhi i tētahi i muri ake nei.
Kua tae te tepe taumata wātea
Kua whakamahia e koe ōna pūāhua wātea 5,000 i ia rā. Whakatū i tētahi kāwanatanga wātea kia whiwhi ai ki ngā pūāhua tāpiri 15,000 tae atu ki ngā pūāhua wātea 10,000 i ia marama.
