621 lines
24 KiB
Python
621 lines
24 KiB
Python
import os
|
|
import uuid
|
|
import time
|
|
import torchaudio
|
|
from typing import Optional
|
|
import folder_paths
|
|
import comfy.utils
|
|
import random
|
|
import requests
|
|
import time
|
|
import json
|
|
import os
|
|
from comfy_api.latest import IO,UI
|
|
|
|
class IndexTTSClient:
|
|
def __init__(self, base_url="http://localhost:5000",output_path="outputs"):
|
|
self.base_url = base_url
|
|
self.output_path = output_path
|
|
|
|
def initialize_tts(self):
|
|
"""调用接口初始化TTS模型"""
|
|
if self.is_model_loaded():
|
|
return True
|
|
json_data = requests.post(f"{self.base_url}/initialize").json()
|
|
if json_data.get("status") == "success":
|
|
return True
|
|
else:
|
|
return False
|
|
|
|
def unload_model(self):
|
|
"""调用接口卸载模型"""
|
|
if not self.is_model_loaded():
|
|
return True
|
|
json_data = requests.post(f"{self.base_url}/unload").json()
|
|
if json_data.get("status") == "success":
|
|
return True
|
|
else:
|
|
return False
|
|
|
|
def is_model_loaded(self):
|
|
"""检查模型是否已加载"""
|
|
json_data = requests.get(f"{self.base_url}/is_loaded").json()
|
|
return json_data.get("is_loaded")
|
|
def get_task_status(self):
|
|
"""获取当前任务状态"""
|
|
try:
|
|
response = requests.get(f"{self.base_url}/task/status")
|
|
return response.json()
|
|
except Exception as e:
|
|
return {"error": str(e)}
|
|
|
|
def get_task_history(self, limit=10):
|
|
"""获取任务历史"""
|
|
try:
|
|
response = requests.get(f"{self.base_url}/task/history", params={"limit": limit})
|
|
return response.json()
|
|
except Exception as e:
|
|
return {"error": str(e)}
|
|
|
|
def get_task_result(self, task_id):
|
|
"""获取任务结果"""
|
|
try:
|
|
response = requests.get(f"{self.base_url}/task/result/{task_id}")
|
|
if response.status_code == 200:
|
|
# 保存音频文件
|
|
filename = os.path.join(self.output_path,f"output_{task_id}.wav")
|
|
with open(filename, 'wb') as f:
|
|
f.write(response.content)
|
|
return {"status": "success", "filename": filename}
|
|
else:
|
|
return response.json()
|
|
except Exception as e:
|
|
return {"error": str(e)}
|
|
|
|
def preview_segments(self, text, max_tokens=100):
|
|
"""获取文本分句预览"""
|
|
try:
|
|
data = {
|
|
"text": text,
|
|
"max_tokens_per_segment": max_tokens
|
|
}
|
|
response = requests.post(f"{self.base_url}/preview_segments", json=data)
|
|
return response.json()
|
|
except Exception as e:
|
|
return {"error": str(e)}
|
|
|
|
def generate_speech(self, text, prompt_audio_path=None, emo_ref_audio_path=None,
|
|
emo_control_method=0, emo_weight=1.0, emo_text="", emo_random=False,
|
|
emotion_vectors=None, **kwargs):
|
|
"""生成语音
|
|
|
|
参数:
|
|
text: 要转换为语音的文本
|
|
prompt_audio_path: 音色参考音频文件路径
|
|
emo_ref_audio_path: 情感参考音频文件路径
|
|
emo_control_method: 情感控制方式 (0-3)
|
|
emo_weight: 情感权重
|
|
emo_text: 情感描述文本
|
|
emo_random: 是否启用情感随机采样
|
|
emotion_vectors: 情感向量列表 [vec1, vec2, ..., vec8]
|
|
**kwargs: 其他高级参数
|
|
"""
|
|
# 检查是否有任务正在进行
|
|
status = self.get_task_status()
|
|
if "current_task" in status and status["current_task"] is not None:
|
|
return {"error": "已有任务正在处理中，请稍后再试"}
|
|
|
|
# 准备表单数据
|
|
data = {
|
|
"text": text,
|
|
"emo_control_method": str(emo_control_method),
|
|
"emo_weight": str(emo_weight),
|
|
"emo_text": emo_text,
|
|
"emo_random": "true" if emo_random else "false",
|
|
}
|
|
|
|
# 添加情感向量
|
|
if emotion_vectors and len(emotion_vectors) == 8:
|
|
for i, vec in enumerate(emotion_vectors, 1):
|
|
data[f"vec{i}"] = str(vec)
|
|
|
|
# 添加高级参数
|
|
advanced_params = {
|
|
"do_sample": True,
|
|
"top_p": 0.8,
|
|
"top_k": 30,
|
|
"temperature": 0.8,
|
|
"length_penalty": 0.0,
|
|
"num_beams": 3,
|
|
"repetition_penalty": 10.0,
|
|
"max_mel_tokens": 1500,
|
|
"max_text_tokens_per_segment": 100
|
|
}
|
|
advanced_params.update(kwargs)
|
|
|
|
for key, value in advanced_params.items():
|
|
if isinstance(value, bool):
|
|
data[key] = "true" if value else "false"
|
|
else:
|
|
data[key] = str(value)
|
|
|
|
# 准备文件
|
|
files = {}
|
|
if prompt_audio_path and os.path.exists(prompt_audio_path):
|
|
files["prompt_audio"] = open(prompt_audio_path, 'rb')
|
|
|
|
if emo_ref_audio_path and os.path.exists(emo_ref_audio_path):
|
|
files["emo_ref_audio"] = open(emo_ref_audio_path, 'rb')
|
|
|
|
try:
|
|
# 发送请求
|
|
response = requests.post(f"{self.base_url}/generate", data=data, files=files)
|
|
result = response.json()
|
|
|
|
# 关闭文件
|
|
for file in files.values():
|
|
file.close()
|
|
|
|
return result
|
|
except Exception as e:
|
|
# 确保文件被关闭
|
|
for file in files.values():
|
|
file.close()
|
|
return {"error": str(e)}
|
|
|
|
def wait_for_task_completion(self, task_id, check_interval=2, timeout=300):
|
|
"""等待任务完成
|
|
|
|
参数:
|
|
task_id: 任务ID
|
|
check_interval: 检查间隔(秒)
|
|
timeout: 超时时间(秒)
|
|
"""
|
|
start_time = time.time()
|
|
while time.time() - start_time < timeout:
|
|
status = self.get_task_status()
|
|
# 检查任务是否已完成
|
|
if status.get("current_task") is None:
|
|
# 检查任务历史
|
|
history = self.get_task_history(limit=1)
|
|
if "tasks" in history:
|
|
for task in history["tasks"]:
|
|
if task.get("id") == task_id:
|
|
if task.get("status") == "completed":
|
|
return {"status": "completed", "task": task}
|
|
elif task.get("status") == "failed":
|
|
return {"status": "failed", "error": task.get("error", "未知错误")}
|
|
|
|
# 任务仍在处理中
|
|
print(f"任务 {task_id} 仍在处理中...")
|
|
time.sleep(check_interval)
|
|
|
|
return {"status": "timeout", "error": "任务处理超时"}
|
|
|
|
class LamIndexTTS2AdvancedParams:
|
|
def __init__(self):
|
|
self.output_dir = folder_paths.get_output_directory()
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"do_sample": ([True, False], {"default": True,"tooltip": "是否使用采样"}),
|
|
"temperature": ("FLOAT", {"default": 0.8, "min": 0.1, "max":2.0, "step": 0.1,"tooltip": "采样温度"}),
|
|
"top_p": ("FLOAT", {"default": 0.8, "min": 0.0, "max":1.0, "step": 0.01,"tooltip": "采样概率"}),
|
|
"top_k": ("INT", {"default": 30, "min": 0, "max":100, "step": 1,"tooltip": "采样数量"}),
|
|
"num_beams": ("INT", {"default": 3, "min": 1, "max":10, "step": 1,"tooltip": "num_beams"}),
|
|
"repetition_penalty": ("FLOAT", {"default": 10.0, "min": 0.1, "max":20.0, "step": 0.1,"tooltip": "repetition_penalty"}),
|
|
"length_penalty": ("FLOAT", {"default": 0.0, "min": -2.0, "max":2.0, "step": 0.1,"tooltip": "length_penalty"}),
|
|
"max_mel_tokens": ("INT", {"default": 1500, "min": 50, "max":1815, "step": 10,"tooltip": "生成Token最大数量，过小导致音频被截断"}),
|
|
#"initial_value": ("STRING",{"default": 100,"min": 20, "max":600, "step": 10,"tooltip": "分句最大Token数"}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("ADVANCED_PARAMS", )
|
|
RETURN_NAMES = ("高级参数",)
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
|
|
def process(self,**kwargs):
|
|
return (kwargs,)
|
|
|
|
class LamIndexTTS2UnloadModel:
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"server": ("STRING", {"default": "http://localhost:5000", "tooltip": "请勿修改"}),
|
|
}
|
|
}
|
|
RETURN_TYPES = ("BOOL", )
|
|
RETURN_NAMES = ("isOk", )
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
def process(self,server:str):
|
|
client = IndexTTSClient(server)
|
|
result = client.unload_model()
|
|
return (result,)
|
|
|
|
class LamIndexTTS2Node0():
|
|
def __init__(self):
|
|
self.audio_dir = folder_paths.get_output_directory()
|
|
self.output_dir = folder_paths.get_temp_directory()
|
|
self.type = "temp"
|
|
self.prefix_append = ""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"server": ("STRING", {"default": "http://localhost:5000", "tooltip": "请勿修改"}),
|
|
"audio": ("AUDIO,STRING", {"default": "", "forceInput": True, "tooltip": "参考音频地址"}),
|
|
"text": ("STRING",{"default": "","multiline": True,"tooltip": "合成内容"}),
|
|
},
|
|
"optional": {
|
|
"params": ("ADVANCED_PARAMS", ),
|
|
"audio_dir": ("STRING", {"default": "", "forceInput": True, "tooltip": "文件夹"}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("STRING", "STRING",)
|
|
RETURN_NAMES = ("audio_path","task_id",)
|
|
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
|
|
def process(self,server:str,audio,text:str,params={},audio_dir=""):
|
|
if isinstance(audio, str):
|
|
audio_path = audio
|
|
else:
|
|
# 生成临时音频文件
|
|
results=UI.AudioSaveHelper.save_audio(
|
|
audio,
|
|
filename_prefix="ComfyUI",
|
|
folder_type=IO.FolderType.temp,
|
|
cls=None,
|
|
format="mp3")
|
|
paths=[]
|
|
for i in range(len(results)):
|
|
subfolder=results[i]['subfolder']
|
|
if subfolder:
|
|
path=os.path.join(self.output_dir, results[i]['subfolder'],results[i]['filename'])
|
|
else:
|
|
path=os.path.join(self.output_dir, results[i]['filename'])
|
|
paths.append(path)
|
|
audio_path = paths[0]
|
|
|
|
if audio_dir !="":
|
|
audio_dir=os.path.join(self.audio_dir,audio_dir)
|
|
else:
|
|
audio_dir=self.audio_dir
|
|
client = IndexTTSClient(server,audio_dir)
|
|
is_ok = client.is_model_loaded()
|
|
if not is_ok:
|
|
client.initialize_tts()
|
|
|
|
result = client.generate_speech(text=text,prompt_audio_path=audio_path
|
|
,emo_control_method=0 # 使用音色参考音频相同的情感
|
|
,**params)
|
|
if "error" in result:
|
|
raise Exception(result["error"])
|
|
else:
|
|
task_id = result.get("task_id")
|
|
print(f"任务已提交，任务ID: {task_id}")
|
|
|
|
# 等待任务完成
|
|
wait_result = client.wait_for_task_completion(task_id)
|
|
print(f"任务状态: {wait_result['status']}")
|
|
|
|
if wait_result["status"] == "completed":
|
|
# 获取任务结果
|
|
result = client.get_task_result(task_id)
|
|
if "filename" in result:
|
|
print(f"音频已保存到: {result['filename']}")
|
|
return (result["filename"],task_id,)
|
|
|
|
print(f"获取结果失败: {wait_result}")
|
|
raise Exception("获取结果失败")
|
|
|
|
|
|
class LamIndexTTS2Node1():
|
|
def __init__(self):
|
|
self.audio_dir = folder_paths.get_output_directory()
|
|
self.output_dir = folder_paths.get_temp_directory()
|
|
self.type = "temp"
|
|
self.prefix_append = ""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"server": ("STRING", {"default": "http://localhost:5000", "tooltip": "请勿修改"}),
|
|
"audio": ("AUDIO,STRING", {"default": "", "forceInput": True, "tooltip": "音色参考音频地址"}),
|
|
"ref_audio": ("AUDIO,STRING", {"default": "", "forceInput": True, "tooltip": "情感参考音频地址"}),
|
|
"text": ("STRING",{"default": "","multiline": True,"tooltip": "合成内容"}),
|
|
},
|
|
"optional": {
|
|
"params": ("ADVANCED_PARAMS", ),
|
|
"audio_dir": ("STRING", {"default": "", "forceInput": True, "tooltip": "文件夹"}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("STRING", "STRING",)
|
|
RETURN_NAMES = ("audio_path","task_id",)
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
|
|
def process(self,server:str,audio,ref_audio,text:str,params={},audio_dir=""):
|
|
if isinstance(audio, str):
|
|
audio_path = audio
|
|
else:
|
|
results=UI.AudioSaveHelper.save_audio(
|
|
audio,
|
|
filename_prefix="ComfyUI",
|
|
folder_type=IO.FolderType.temp,
|
|
cls=None,
|
|
format="mp3")
|
|
paths=[]
|
|
for i in range(len(results)):
|
|
subfolder=results[i]['subfolder']
|
|
if subfolder:
|
|
path=os.path.join(self.output_dir, results[i]['subfolder'],results[i]['filename'])
|
|
else:
|
|
path=os.path.join(self.output_dir, results[i]['filename'])
|
|
paths.append(path)
|
|
audio_path = paths[0]
|
|
|
|
if isinstance(ref_audio, str):
|
|
ref_audio_path = ref_audio
|
|
else:
|
|
results=UI.AudioSaveHelper.save_audio(
|
|
audio,
|
|
filename_prefix="ComfyUI",
|
|
folder_type=IO.FolderType.temp,
|
|
cls=None,
|
|
format="mp3")
|
|
paths=[]
|
|
for i in range(len(results)):
|
|
subfolder=results[i]['subfolder']
|
|
if subfolder:
|
|
path=os.path.join(self.output_dir, results[i]['subfolder'],results[i]['filename'])
|
|
else:
|
|
path=os.path.join(self.output_dir, results[i]['filename'])
|
|
paths.append(path)
|
|
ref_audio_path = paths[0]
|
|
|
|
if audio_dir !="":
|
|
audio_dir=os.path.join(self.audio_dir,audio_dir)
|
|
else:
|
|
audio_dir=self.audio_dir
|
|
client = IndexTTSClient(server,audio_dir)
|
|
is_ok = client.is_model_loaded()
|
|
if not is_ok:
|
|
client.initialize_tts()
|
|
|
|
result = client.generate_speech(text=text,prompt_audio_path=audio_path
|
|
,emo_ref_audio_path=ref_audio_path,emo_control_method=1 # 使用音色参考音频相同的情感
|
|
,**params
|
|
)
|
|
if "error" in result:
|
|
raise Exception(result["error"])
|
|
else:
|
|
task_id = result.get("task_id")
|
|
print(f"任务已提交，任务ID: {task_id}")
|
|
|
|
# 等待任务完成
|
|
wait_result = client.wait_for_task_completion(task_id)
|
|
print(f"任务状态: {wait_result['status']}")
|
|
|
|
if wait_result["status"] == "completed":
|
|
# 获取任务结果
|
|
result = client.get_task_result(task_id)
|
|
if "filename" in result:
|
|
print(f"音频已保存到: {result['filename']}")
|
|
return (result["filename"],task_id,)
|
|
|
|
print(f"获取结果失败: {wait_result}")
|
|
raise Exception("获取结果失败")
|
|
|
|
class LamIndexTTS2Node2():
|
|
def __init__(self):
|
|
self.audio_dir = folder_paths.get_output_directory()
|
|
self.output_dir = folder_paths.get_temp_directory()
|
|
self.type = "temp"
|
|
self.prefix_append = ""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"server": ("STRING", {"default": "http://localhost:5000", "tooltip": "请勿修改"}),
|
|
"audio": ("AUDIO,STRING", {"default": "", "forceInput": True, "tooltip": "参考音频地址"}),
|
|
"vec1": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "喜"}),
|
|
"vec2": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "怒"}),
|
|
"vec3": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "哀"}),
|
|
"vec4": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "惧"}),
|
|
"vec5": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "厌恶"}),
|
|
"vec6": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "低落"}),
|
|
"vec7": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "惊喜"}),
|
|
"vec8": ("FLOAT", {"default": 0.0, "min": 0.0, "max":1.4, "step": 0.05,"tooltip": "平静"}),
|
|
"text": ("STRING",{"default": "","multiline": True,"tooltip": "合成内容"}),
|
|
},
|
|
"optional": {
|
|
"params": ("ADVANCED_PARAMS", ),
|
|
"audio_dir": ("STRING", {"default": "", "forceInput": True, "tooltip": "文件夹"}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("STRING", "STRING",)
|
|
RETURN_NAMES = ("audio_path","task_id",)
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
|
|
def process(self,server:str,audio,vec1:float,vec2:float,vec3:float,vec4:float,vec5:float,vec6:float,vec7:float,vec8:float,text:str,params={},audio_dir=""):
|
|
task_id = str(uuid.uuid4())
|
|
if isinstance(audio, str):
|
|
audio_path = audio
|
|
else:
|
|
results=UI.AudioSaveHelper.save_audio(
|
|
audio,
|
|
filename_prefix="ComfyUI",
|
|
folder_type=IO.FolderType.temp,
|
|
cls=None,
|
|
format="mp3")
|
|
paths=[]
|
|
for i in range(len(results)):
|
|
subfolder=results[i]['subfolder']
|
|
if subfolder:
|
|
path=os.path.join(self.output_dir, results[i]['subfolder'],results[i]['filename'])
|
|
else:
|
|
path=os.path.join(self.output_dir, results[i]['filename'])
|
|
paths.append(path)
|
|
audio_path = paths[0]
|
|
|
|
if audio_dir !="":
|
|
audio_dir=os.path.join(self.audio_dir,audio_dir)
|
|
else:
|
|
audio_dir=self.audio_dir
|
|
client = IndexTTSClient(server,audio_dir)
|
|
emotion_vectors=[vec1,vec2,vec3,vec4,vec5,vec6,vec7,vec8]
|
|
is_ok = client.is_model_loaded()
|
|
if not is_ok:
|
|
client.initialize_tts()
|
|
|
|
result = client.generate_speech(text=text,prompt_audio_path=audio_path
|
|
,emotion_vectors=emotion_vectors,emo_control_method=2 # 使用音色参考音频相同的情感
|
|
,**params
|
|
)
|
|
if "error" in result:
|
|
raise Exception(result["error"])
|
|
else:
|
|
task_id = result.get("task_id")
|
|
print(f"任务已提交，任务ID: {task_id}")
|
|
|
|
# 等待任务完成
|
|
wait_result = client.wait_for_task_completion(task_id)
|
|
print(f"任务状态: {wait_result['status']}")
|
|
|
|
if wait_result["status"] == "completed":
|
|
# 获取任务结果
|
|
result = client.get_task_result(task_id)
|
|
if "filename" in result:
|
|
print(f"音频已保存到: {result['filename']}")
|
|
return (result["filename"],task_id,)
|
|
|
|
print(f"获取结果失败: {wait_result}")
|
|
raise Exception("获取结果失败")
|
|
|
|
class LamIndexTTS2Node3():
|
|
def __init__(self):
|
|
self.audio_dir = folder_paths.get_output_directory()
|
|
self.output_dir = folder_paths.get_temp_directory()
|
|
self.type = "temp"
|
|
self.prefix_append = ""
|
|
@classmethod
|
|
def INPUT_TYPES(s):
|
|
return {
|
|
"required": {
|
|
"server": ("STRING", {"default": "http://localhost:5000", "tooltip": "请勿修改"}),
|
|
"audio": ("AUDIO,STRING", {"default": "", "forceInput": True, "tooltip": "参考音频地址"}),
|
|
"text": ("STRING",{"default": "","multiline": True,"tooltip": "合成内容"}),
|
|
"emo_text": ("STRING", {"default": "", "tooltip": "情感描述文本"}),
|
|
},
|
|
"optional": {
|
|
"params": ("ADVANCED_PARAMS", ),
|
|
"audio_dir": ("STRING", {"default": "", "forceInput": True, "tooltip": "文件夹"}),
|
|
}
|
|
}
|
|
|
|
RETURN_TYPES = ("STRING", "STRING",)
|
|
RETURN_NAMES = ("audio_path","task_id",)
|
|
FUNCTION = "process"
|
|
|
|
OUTPUT_NODE = False
|
|
|
|
CATEGORY = "lam"
|
|
|
|
def process(self,server:str,audio,text:str,emo_text:str,params={},audio_dir=""):
|
|
task_id = str(uuid.uuid4())
|
|
if isinstance(audio, str):
|
|
audio_path = audio
|
|
else:
|
|
results=UI.AudioSaveHelper.save_audio(
|
|
audio,
|
|
filename_prefix="ComfyUI",
|
|
folder_type=IO.FolderType.temp,
|
|
cls=None,
|
|
format="mp3")
|
|
paths=[]
|
|
for i in range(len(results)):
|
|
subfolder=results[i]['subfolder']
|
|
if subfolder:
|
|
path=os.path.join(self.output_dir, results[i]['subfolder'],results[i]['filename'])
|
|
else:
|
|
path=os.path.join(self.output_dir, results[i]['filename'])
|
|
paths.append(path)
|
|
audio_path = paths[0]
|
|
|
|
if audio_dir !="":
|
|
audio_dir=os.path.join(self.audio_dir,audio_dir)
|
|
else:
|
|
audio_dir=self.audio_dir
|
|
client = IndexTTSClient(server,audio_dir)
|
|
is_ok = client.is_model_loaded()
|
|
if not is_ok:
|
|
client.initialize_tts()
|
|
|
|
result = client.generate_speech(text=text,prompt_audio_path=audio_path
|
|
,emo_text=emo_text,emo_control_method=3 # 使用音色参考音频相同的情感
|
|
,**params
|
|
)
|
|
if "error" in result:
|
|
raise Exception(result["error"])
|
|
else:
|
|
task_id = result.get("task_id")
|
|
print(f"任务已提交，任务ID: {task_id}")
|
|
|
|
# 等待任务完成
|
|
wait_result = client.wait_for_task_completion(task_id)
|
|
print(f"任务状态: {wait_result['status']}")
|
|
|
|
if wait_result["status"] == "completed":
|
|
# 获取任务结果
|
|
result = client.get_task_result(task_id)
|
|
if "filename" in result:
|
|
print(f"音频已保存到: {result['filename']}")
|
|
return (result["filename"],task_id,)
|
|
|
|
print(f"获取结果失败: {wait_result}")
|
|
raise Exception("获取结果失败")
|
|
|
|
|
|
|
|
NODE_CLASS_MAPPINGS = {
|
|
"LamIndexTTS2AdvancedParams": LamIndexTTS2AdvancedParams,
|
|
"LamIndexTTS2UnloadModel": LamIndexTTS2UnloadModel,
|
|
"LamIndexTTS2Node0": LamIndexTTS2Node0,
|
|
"LamIndexTTS2Node1": LamIndexTTS2Node1,
|
|
"LamIndexTTS2Node2": LamIndexTTS2Node2,
|
|
"LamIndexTTS2Node3": LamIndexTTS2Node3,
|
|
}
|
|
|
|
NODE_DISPLAY_NAME_MAPPINGS = {
|
|
"LamIndexTTS2AdvancedParams": "IndexTTS2高级参数",
|
|
"LamIndexTTS2UnloadModel": "IndexTTS2-卸载模型",
|
|
"LamIndexTTS2Node0": "IndexTTS2-音色情感同时参考",
|
|
"LamIndexTTS2Node1": "IndexTTS2-音色情感分别参考",
|
|
"LamIndexTTS2Node2": "IndexTTS2-音色参考+情感向量",
|
|
"LamIndexTTS2Node3": "IndexTTS2-音色参考+情感描述",
|
|
} |