app.py

import gradio as gr
from melo.api import TTS
import torch

# 设定设备
device = 'cuda:0' if torch.cuda.is_available() else 'cpu'

# 加载模型
model = TTS(language='ZH', device=device)
speaker_ids = model.hps.data.spk2id

def text_to_speech(text, speed):
    # 从模型生成音频
    output_path = 'output.wav'
    model.tts_to_file(text, speaker_ids['ZH'], output_path, speed=speed)
    return output_path

# 创建 Gradio 界面
iface = gr.Interface(
    fn=text_to_speech,
    inputs=[
        gr.Textbox(label="输入中文文本", value="我最近在学习机器学习，希望能够在未来的人工智能领域有所建树。"),
        gr.Slider(minimum=0.5, maximum=1.5, step=0.1, value=1.0, label="语速")
    ],
    outputs=gr.Audio(label="生成的语音"),
    title="中文文本转语音",
    description="输入一段中文文本，然后点击“Submit”按钮来生成语音。"
)

# 启动应用
iface.launch()