免费文字转语音助手 python+edge_tts+FFMPEG

需要安装edge_tts和FFMPEG

pip install edge_tts

FFMPEG安装参考https://blog.csdn.net/Natsuago/article/details/143231558

import edge_tts
import asyncio
import os
import re
import tkinter as tk
from tkinter import ttk, messagebox, filedialog
import threading
import sys

# ===================== 配置FFmpeg路径（自动检测或手动配置） =====================
# 优先使用系统环境变量中的ffmpeg，如果没有则使用指定路径
FFMPEG_PATH = os.environ.get("FFMPEG_PATH", r"D:\Program Files\ffmpeg-7.1.1-essentials_build\bin\ffmpeg.exe")
FFPROBE_PATH = os.environ.get("FFPROBE_PATH", r"D:\Program Files\ffmpeg-7.1.1-essentials_build\bin\ffprobe.exe")

# 验证FFmpeg路径
def check_ffmpeg():
    """检查FFmpeg是否可用"""
    if not os.path.exists(FFMPEG_PATH):
        messagebox.showwarning("警告", f"FFmpeg未找到：{FFMPEG_PATH}\n请检查路径配置！")
        return False
    return True

# 配置FFmpeg环境变量
os.environ["FFMPEG_PATH"] = FFMPEG_PATH
os.environ["FFPROBE_PATH"] = FFPROBE_PATH

# ===================== 扩展版语音音色列表 =====================
VOICE_OPTIONS = {
    # ==================== 中文普通话 - 女声 ====================
    "【中文】晓晓 (默认)": "zh-CN-XiaoxiaoNeural",
    "【中文】小艺": "zh-CN-XiaoyiNeural",
    "【中文】小燕": "zh-CN-XiaoyanNeural",
    "【中文】小云": "zh-CN-XiaoyunNeural",
    "【中文】小梦": "zh-CN-XiaomengNeural",
    "【中文】小希": "zh-CN-XiaoxiNeural",
    "【中文】小雅": "zh-CN-XiaoyaNeural",
    "【中文】小宁": "zh-CN-XiaoningNeural",
    "【中文】小莉": "zh-CN-LiNeural",
    "【中文】小月": "zh-CN-YueNeural",
    
    # ==================== 中文普通话 - 男声 ====================
    "【中文】云希": "zh-CN-YunxiNeural",
    "【中文】云健": "zh-CN-YunjianNeural",
    "【中文】云扬": "zh-CN-YunyangNeural",
    "【中文】晓辰": "zh-CN-XiaochenNeural",
    "【中文】晓宇": "zh-CN-XiaoyuNeural",
    "【中文】小伟": "zh-CN-WeiNeural",
    "【中文】小鹏": "zh-CN-PengNeural",
    "【中文】小川": "zh-CN-ChuanNeural",
    
    # ==================== 中文方言 ====================
    "【方言】粤语-小玲": "zh-HK-XiaolingNeural",
    "【方言】粤语-大雄": "zh-HK-DaanNeural",
    "【方言】台湾话-晓萱": "zh-TW-HsiaoHsuanNeural",
    "【方言】台湾话-晓东": "zh-TW-HsiaoTungNeural",
    
    # ==================== 英文 - 美式 ====================
    "【英文】Aria (女声)": "en-US-AriaNeural",
    "【英文】Jenny (女声)": "en-US-JennyNeural",
    "【英文】Michelle (女声)": "en-US-MichelleNeural",
    "【英文】Sara (女声)": "en-US-SaraNeural",
    "【英文】Brandon (男声)": "en-US-BrandonNeural",
    "【英文】Christopher (男声)": "en-US-ChristopherNeural",
    "【英文】Eric (男声)": "en-US-EricNeural",
    "【英文】Guy (男声)": "en-US-GuyNeural",
    
    # ==================== 英文 - 英式 ====================
    "【英文】Libby (英式女声)": "en-GB-LibbyNeural",
    "【英文】Maisie (英式女声)": "en-GB-MaisieNeural",
    "【英文】Ryan (英式男声)": "en-GB-RyanNeural",
    "【英文】Thomas (英式男声)": "en-GB-ThomasNeural",
    
    # ==================== 其他特色音色 ====================
    "【英文】小女孩-Emma": "en-US-EmmaNeural",
    "【英文】小男孩-Brian": "en-US-BrianNeural",
    "【中文】情感女声-晓晨": "zh-CN-XiaochenNeural",
    "【中文】轻柔女声-晓雪": "zh-CN-XiaoxueNeural"
}

def get_valid_filename(text):
    """将文本转换为合法的文件名"""
    # 移除Windows非法字符
    filename = re.sub(r'[\\/:*?"<>|]', '', text)
    # 移除空白字符并限制长度
    filename = filename.strip()
    if len(filename) > 50:
        filename = filename[:47] + "..."
    if not filename:
        filename = "默认语音"
    return f"{filename}.mp3"

async def generate_voice(text, voice, rate, volume, pitch, save_path):
    """异步生成语音文件"""
    try:
        # 创建Communicate实例
        communicate = edge_tts.Communicate(
            text=text,
            voice=voice,
            rate=rate,
            volume=volume,
            pitch=pitch
        )
        
        # 生成并保存文件
        await communicate.save(save_path)
        return True
    except Exception as e:
        return str(e)

def run_async_task(loop, coro):
    """在指定事件循环中运行异步任务"""
    return loop.run_until_complete(coro)

def on_generate():
    """生成语音按钮点击事件（优化版）"""
    # 检查FFmpeg
    if not check_ffmpeg():
        return
    
    # 获取输入文本
    text = text_input.get("1.0", tk.END).strip()
    if not text:
        messagebox.showwarning("警告", "请输入要转换的文本！")
        return
    
    # 获取选中的音色
    voice_name = voice_var.get()
    voice_code = VOICE_OPTIONS[voice_name]
    
    # 获取语速/音量/音调（转为整数，确保格式正确）
    try:
        rate_value = int(rate_slider.get())
        volume_value = int(volume_slider.get())
        pitch_value = int(pitch_slider.get())
        
        rate = f"{rate_value:+d}%"  # 确保有正负号
        volume = f"{volume_value:+d}%"
        pitch = f"{pitch_value:+d}Hz"
    except:
        messagebox.showerror("错误", "参数转换失败，请检查滑块设置！")
        return
    
    # 生成合法文件名
    filename = get_valid_filename(text[:100])  # 只取前100个字符作为文件名
    # 选择保存路径
    save_path = filedialog.asksaveasfilename(
        defaultextension=".mp3",
        initialfile=filename,
        filetypes=[("MP3文件", "*.mp3"), ("所有文件", "*.*")],
        title="保存语音文件"
    )
    if not save_path:
        return
    
    # 禁用按钮防止重复点击
    generate_btn.config(state="disabled")
    generate_btn.config(text="正在生成...")
    root.update()
    
    try:
        # 在新线程中运行异步任务，避免GUI冻结
        def generate_thread():
            loop = asyncio.new_event_loop()
            asyncio.set_event_loop(loop)
            result = loop.run_until_complete(generate_voice(
                text, voice_code, rate, volume, pitch, save_path
            ))
            
            # 回到主线程更新UI
            def show_result():
                generate_btn.config(state="normal")
                generate_btn.config(text="生成语音文件")
                if result is True:
                    messagebox.showinfo("成功", f"语音文件已生成：\n{save_path}")
                    # 可选：打开文件所在目录
                    if messagebox.askyesno("提示", "是否打开文件所在目录？"):
                        os.startfile(os.path.dirname(save_path))
                else:
                    messagebox.showerror("错误", f"生成失败：\n{result}")
            
            root.after(0, show_result)
        
        # 启动生成线程
        thread = threading.Thread(target=generate_thread, daemon=True)
        thread.start()
        
    except Exception as e:
        generate_btn.config(state="normal")
        generate_btn.config(text="生成语音文件")
        messagebox.showerror("错误", f"程序异常：{str(e)}")

def clear_text():
    """清空文本输入框"""
    text_input.delete("1.0", tk.END)

def load_text_file():
    """从文件加载文本"""
    file_path = filedialog.askopenfilename(
        filetypes=[("文本文件", "*.txt"), ("所有文件", "*.*")],
        title="选择文本文件"
    )
    if file_path:
        try:
            with open(file_path, "r", encoding="utf-8") as f:
                text = f.read()
            text_input.delete("1.0", tk.END)
            text_input.insert("1.0", text)
        except Exception as e:
            messagebox.showerror("错误", f"读取文件失败：{str(e)}")

# ===================== 创建GUI界面 =====================
def create_gui():
    global root, text_input, voice_var, rate_slider, volume_slider, pitch_slider, generate_btn
    
    root = tk.Tk()
    root.title("Edge TTS 语音生成工具 v2.2")
    root.geometry("700x600")
    root.resizable(True, True)
    
    # 设置字体
    default_font = ("微软雅黑", 9)
    root.option_add("*Font", default_font)
    
    # 1. 标题和说明
    title_label = ttk.Label(root, text="Edge TTS 语音生成工具 (扩展音色版)", font=("微软雅黑", 12, "bold"))
    title_label.pack(pady=10)
    
    # 2. 文本输入区域（带滚动条）
    frame_text = ttk.Frame(root)
    frame_text.pack(padx=20, pady=5, fill=tk.BOTH, expand=True)
    
    ttk.Label(frame_text, text="请输入要转换的文本：").anchor(tk.W)
    text_scroll = ttk.Scrollbar(frame_text)
    text_input = tk.Text(frame_text, width=80, height=10, wrap=tk.WORD,
                         yscrollcommand=text_scroll.set, font=("微软雅黑", 10))
    text_scroll.config(command=text_input.yview)
    
    text_input.pack(side=tk.LEFT, fill=tk.BOTH, expand=True)
    text_scroll.pack(side=tk.RIGHT, fill=tk.Y)
    
    # 文本操作按钮
    frame_text_buttons = ttk.Frame(root)
    frame_text_buttons.pack(padx=20, pady=5, fill=tk.X)
    
    ttk.Button(frame_text_buttons, text="清空文本", command=clear_text).pack(side=tk.LEFT, padx=5)
    ttk.Button(frame_text_buttons, text="加载文本文件", command=load_text_file).pack(side=tk.LEFT, padx=5)
    
    # 3. 参数设置区域
    frame_params = ttk.LabelFrame(root, text="语音参数设置")
    frame_params.pack(padx=20, pady=10, fill=tk.X)
    
    # 3.1 音色选择（加宽下拉框以显示完整音色名称）
    frame_voice = ttk.Frame(frame_params)
    frame_voice.pack(padx=20, pady=8, fill=tk.X)
    
    ttk.Label(frame_voice, text="语音音色：", width=10).pack(side=tk.LEFT)
    voice_var = tk.StringVar(value=list(VOICE_OPTIONS.keys())[0])
    voice_combobox = ttk.Combobox(frame_voice, textvariable=voice_var, 
                                  values=list(VOICE_OPTIONS.keys()), state="readonly", width=45)
    voice_combobox.pack(side=tk.LEFT, padx=5)
    
    # 3.2 语速调节（默认值设置为10）
    frame_rate = ttk.Frame(frame_params)
    frame_rate.pack(padx=20, pady=5, fill=tk.X)
    
    ttk.Label(frame_rate, text="语速：", width=10).pack(side=tk.LEFT)
    rate_slider = ttk.Scale(frame_rate, from_=-100, to=100, orient=tk.HORIZONTAL, length=450)
    rate_slider.set(10)  # 关键修改：默认语速设置为10
    rate_slider.pack(side=tk.LEFT, padx=5)
    rate_label = ttk.Label(frame_rate, text="10%", width=5)  # 默认显示10%
    rate_label.pack(side=tk.LEFT)
    
    def update_rate_label(event):
        rate_label.config(text=f"{int(rate_slider.get())}%")
    rate_slider.bind("<Motion>", update_rate_label)
    rate_slider.bind("<ButtonRelease>", update_rate_label)
    
    # 3.3 音量调节
    frame_volume = ttk.Frame(frame_params)
    frame_volume.pack(padx=20, pady=5, fill=tk.X)
    
    ttk.Label(frame_volume, text="音量：", width=10).pack(side=tk.LEFT)
    volume_slider = ttk.Scale(frame_volume, from_=-100, to=100, orient=tk.HORIZONTAL, length=450)
    volume_slider.set(0)
    volume_slider.pack(side=tk.LEFT, padx=5)
    volume_label = ttk.Label(frame_volume, text="0%", width=5)
    volume_label.pack(side=tk.LEFT)
    
    def update_volume_label(event):
        volume_label.config(text=f"{int(volume_slider.get())}%")
    volume_slider.bind("<Motion>", update_volume_label)
    volume_slider.bind("<ButtonRelease>", update_volume_label)
    
    # 3.4 音调调节
    frame_pitch = ttk.Frame(frame_params)
    frame_pitch.pack(padx=20, pady=5, fill=tk.X)
    
    ttk.Label(frame_pitch, text="音调：", width=10).pack(side=tk.LEFT)
    pitch_slider = ttk.Scale(frame_pitch, from_=-50, to=50, orient=tk.HORIZONTAL, length=450)
    pitch_slider.set(0)
    pitch_slider.pack(side=tk.LEFT, padx=5)
    pitch_label = ttk.Label(frame_pitch, text="0Hz", width=5)
    pitch_label.pack(side=tk.LEFT)
    
    def update_pitch_label(event):
        pitch_label.config(text=f"{int(pitch_slider.get())}Hz")
    pitch_slider.bind("<Motion>", update_pitch_label)
    pitch_slider.bind("<ButtonRelease>", update_pitch_label)
    
    # 4. 生成按钮
    frame_button = ttk.Frame(root)
    frame_button.pack(padx=20, pady=20)
    
    generate_btn = ttk.Button(frame_button, text="生成语音文件", command=on_generate, 
                              width=25, style="Accent.TButton")
    generate_btn.pack(pady=10)
    
    # 5. 状态栏
    status_var = tk.StringVar(value="就绪 | 扩展音色库已加载 | 语速默认值：10% | 支持中文/英文/方言")
    status_bar = ttk.Label(root, textvariable=status_var, relief=tk.SUNKEN, anchor=tk.W)
    status_bar.pack(side=tk.BOTTOM, fill=tk.X)
    
    # 检查FFmpeg
    root.after(100, check_ffmpeg)
    
    root.mainloop()

# 适配Windows系统的异步策略
if sys.platform == 'win32':
    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())

# 启动GUI
if __name__ == "__main__":
    create_gui()