blackteay/GPT-SoVITS
1
0
Fork 0
mirror of https://github.com/RVC-Boss/GPT-SoVITS.git synced 2025-12-31 20:37:03 +08:00
Code Issues Packages Projects Releases Wiki Activity

对齐gpt topk默认采样参数 (#2696)

Browse Source
This commit is contained in:
ChasonJiang 2025-12-19 16:05:36 +08:00 committed by GitHub
parent cc89c3660e
commit abe984395c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
GPT_SoVITS/TTS_infer_pack/TTS.py
View File

@ -1008,7 +1008,7 @@ class TTS:
"aux_ref_audio_paths": [], # list.(optional) auxiliary reference audio paths for multi-speaker tone fusion
"prompt_text": "", # str.(optional) prompt text for the reference audio
"prompt_lang": "", # str.(required) language of the prompt text for the reference audio
"top_k": 5, # int. top k sampling
"top_k": 15, # int. top k sampling
"top_p": 1, # float. top p sampling
"temperature": 1, # float. temperature for sampling
"text_split_method": "cut1", # str. text split method, see text_segmentation_method.py for details.
@ -1039,7 +1039,7 @@ class TTS:
aux_ref_audio_paths: list = inputs.get("aux_ref_audio_paths", [])
prompt_text: str = inputs.get("prompt_text", "")
prompt_lang: str = inputs.get("prompt_lang", "")
top_k: int = inputs.get("top_k", 5)
top_k: int = inputs.get("top_k", 15)
top_p: float = inputs.get("top_p", 1)
temperature: float = inputs.get("temperature", 1)
text_split_method: str = inputs.get("text_split_method", "cut1")

2
GPT_SoVITS/inference_webui_fast.py
View File

@ -385,7 +385,7 @@ with gr.Blocks(title="GPT-SoVITS WebUI", analytics_enabled=False, js=js, css=css
minimum=0.6, maximum=1.65, step=0.05, label="语速", value=1.0, interactive=True
)
with gr.Row():
top_k = gr.Slider(minimum=1, maximum=100, step=1, label=i18n("top_k"), value=5, interactive=True)
top_k = gr.Slider(minimum=1, maximum=100, step=1, label=i18n("top_k"), value=15, interactive=True)
top_p = gr.Slider(minimum=0, maximum=1, step=0.05, label=i18n("top_p"), value=1, interactive=True)
with gr.Row():
temperature = gr.Slider(

8
api_v2.py
View File

@ -27,7 +27,7 @@ POST:
"aux_ref_audio_paths": [], # list.(optional) auxiliary reference audio paths for multi-speaker tone fusion
"prompt_text": "", # str.(optional) prompt text for the reference audio
"prompt_lang": "", # str.(required) language of the prompt text for the reference audio
"top_k": 5, # int. top k sampling
"top_k": 15, # int. top k sampling
"top_p": 1, # float. top p sampling
"temperature": 1, # float. temperature for sampling
"text_split_method": "cut5", # str. text split method, see text_segmentation_method.py for details.
@ -158,7 +158,7 @@ class TTS_Request(BaseModel):
aux_ref_audio_paths: list = None
prompt_lang: str = None
prompt_text: str = ""
top_k: int = 5
top_k: int = 15
top_p: float = 1
temperature: float = 1
text_split_method: str = "cut5"
@ -355,7 +355,7 @@ async def tts_handle(req: dict):
"aux_ref_audio_paths": [], # list.(optional) auxiliary reference audio paths for multi-speaker tone fusion
"prompt_text": "", # str.(optional) prompt text for the reference audio
"prompt_lang": "", # str.(required) language of the prompt text for the reference audio
"top_k": 5, # int. top k sampling
"top_k": 15, # int. top k sampling
"top_p": 1, # float. top p sampling
"temperature": 1, # float. temperature for sampling
"text_split_method": "cut5", # str. text split method, see text_segmentation_method.py for details.
@ -460,7 +460,7 @@ async def tts_get_endpoint(
aux_ref_audio_paths: list = None,
prompt_lang: str = None,
prompt_text: str = "",
top_k: int = 5,
top_k: int = 15,
top_p: float = 1,
temperature: float = 1,
text_split_method: str = "cut5",