Text-to-Speech
English

For use with my fork/draft PR of GPT-SoVITS.


Non-Cherrypicked Examples

Each example was generated in a single attempt using the final checkpoints (Furina_e8_s2136.safetensors & Furina-e15.safetensors) to give an honest look at the models performance. The prompts were randomly generated with ChatGPT.

Generation Settings

Reference Audio: reference_audio.wav
Reference Text: In truth, I know little about becoming a nation's new god, but it will be my honor to guide you all.
No reference mode: Disabled

Speech Rate: 1
Top-K: 20
Min-P: 0.1
Top-P: 1
Temperature: 1
Repetition Penalty: 1.35

Example 1: The flickering lights of the distant city danced like stars on the horizon.

Example 2: A gentle breeze stirred the autumn leaves, carrying with it the scent of rain.

Example 3: Beneath the surface of the ocean, an unseen world teemed with ancient mysteries.

Example 4: In the stillness of the midnight hour, time seemed to pause, holding its breath.

Example 5: The symphony of nature unfolded in the quiet woods, where silence itself seemed alive.

Example 6: Her laughter echoed through the empty halls, a reminder of the joy that once lived there.

Example 7: With each step, the snow crunched beneath her feet, a rhythm as old as winter itself.

Example 8: The stars told stories of forgotten gods and distant worlds, if only one knew how to listen.

Example 9: As the spacecraft rose into the sky, the Earth below became a mere dot in the universe.

Example 10: He stared into the mirror, seeing not his reflection, but the person he might have been.


Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Examples
Unable to determine this model's library. Check the docs .

Model tree for PJMixers-Dev/GPT-SoVITS-Genshin-Impact-Furina

Base model

lj1995/GPT-SoVITS
Finetuned
(1)
this model

Datasets used to train PJMixers-Dev/GPT-SoVITS-Genshin-Impact-Furina