Voice Models v2
EasyAIVoice ↗
32,011 Verified Open Models 28,877 Multi-Pitch Audition Ready 100% Direct Source URLs (HF · Drive · MEGA)
Dora speech model
RVC v2

Dora speech model

Дора, Дарья Шиханова

RVC v2 Russian Peak Quality 5.10 MOS Added Nov 14, 2025 ★ 2 saved
Open Model Weights Vault · Hosted on Hugging Face · 122.7 MB
✓ Verified Direct URL
https://huggingface.co/zAnonymousWizard/DoraSnowie_by_wizard/resolve/main/DoraSnowie.zip?download=true
Download Raw Weights (.zip) ↗
Convert Audio with This Voice in EasyAIVoice
Carries this exact model URL and your selected pitch (0 st) into the cloud studio.
Zero GPU Setup
Audio Decision Loop · Sample Deck

Audition Across 18 Pitch References

Click any semitone cell below to hear how this voice performs on female and male reference vocals and lock your preferred pitch.

★ Sweet Spot: MALE +12 st (5.10 MOS)
FEMALE reference · Pitch 0 st
NISQA Quality 4.75 / 5.0 · Sounds female
Semitone Shift (`-16` to `+16` st) · Bar height indicates NISQA clarity score
Female Ref
Male Ref
Audible Female Audible Male Amber border = Peak NISQA Quality
Raw model title: Dora speech model (Дора, Дарья Шиханова)
CLAP Semantic Analysis · Voice Accords

Sonic Profile at Selected Pitch

Acoustic traits measured directly from the generated output sample. Switch pitches in the Sample Deck to see how timbre shifts.

Age Profile
middle aged adult 45%
elderly 35%
young adult 15%
Music & Vocal Fit
producer tag 44%
music 23%
speaking 10%
Timbre & Texture
villain 45%
cute 28%
anime 18%
Best Use Case
character 66%
gaming 24%
podcasts 6%
Auditory Discovery · Close Alternatives

Voices That Sound Like Dora speech model

Close but not quite the right fit? Audition nearby candidates matched by vocal embedding and timbre.

Luise Meyrink The King of Fighters Maximum Impact · English & Japanese
Nearby voice
MALE · +16 st 4.76 MOS Drive
Harry Potter 1 French dub
Nearby voice
FEMALE · +8 st 4.37 MOS HF
Nearby voice
FEMALE · +8 st 4.61 MOS HF
Nearby voice
FEMALE · +4 st 4.36 MOS HF
Real E-girl Hugging Face
Nearby voice
MALE · +16 st 5.00 MOS HF
Model Documentation & Notes

About Dora speech model (Дора, Дарья Шиханова)

https://huggingface.co/zAnonymousWizard/DoraSnowie_by_wizard/resolve/main/DoraSnowie.zip?download=true Type: RVC v2 Algorithm: rmvpe Epochs: 480 Description: Эпоха 480 была выбрана на основе анализа метрик в TensorBoard В процессе тренировки RVC модели я отслеживал графики обучения в TensorBoard. Именно на 480-й эпохе метрики показали наилучшие результаты, поэтому эта версия модели была выбрана как финальная. Прошу не задавать вопросы по типу: "почему не 1000 эпох?" или "почему не 500?" — количество эпох само по себе не гарантирует качество. Главное — найти баланс и точку, где модель показывает оптимальные метрики без переобучения. В моем случае это 480 эпох. Модель разговорная, но вроде для ИИ каверов тоже неплохо подходит... в любом случае я опубликовал ещё вторую версию Доры, которая как раз-таки для ИИ каверов

Provenance & Open Access
Catalog ID:#32950
Weights Host:Hugging Face
Archive Size:122.7 MB
Rights & Safety: Report / Takedown