Update app.py

app.py

@@ -1,19 +1,33 @@
-import os
-import tempfile
+import gradio as gr
+import edge_tts
 import asyncio
-from moviepy.editor import AudioFileClip, ImageClip, concatenate_videoclips
+import tempfile
+import os
+from moviepy.editor import TextClip, concatenate_videoclips, CompositeVideoClip, AudioFileClip, ImageClip
 from wand.image import Image
-from wand.color import Color
 from wand.drawing import Drawing
-from gtts import gTTS
+from wand.color import Color
+
+# 获取所有可用的语音
+async def get_voices():
+    voices = await edge_tts.list_voices()
+    return {f"{v['ShortName']} - {v['Locale']} ({v['Gender']})": v['ShortName'] for v in voices}
 
-# 文本转语音功能
+# 文字转语音功能
 async def text_to_speech(text, voice, rate, pitch):
-    # 使用 gTTS 生成语音
-    tts = gTTS(text=text, lang=voice)
-    audio_file = os.path.join(tempfile.gettempdir(), "audio.mp3")
-    tts.save(audio_file)
-    return audio_file, None
+    if not text.strip():
+        return None, gr.Warning("Please enter the text to convert.")
+    if not voice:
+        return None, gr.Warning("Please select a voice.")
+    
+    voice_short_name = voice.split(" - ")[0]
+    rate_str = f"{rate:+d}%"
+    pitch_str = f"{pitch:+d}Hz"
+    communicate = edge_tts.Communicate(text, voice_short_name, rate=rate_str, pitch=pitch_str)
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
+        tmp_path = tmp_file.name
+        await communicate.save(tmp_path)
+    return tmp_path, None
 
 # 文字转视频功能
 def text_to_video(text, voice, rate, pitch, video_width, video_height, bg_color, text_color, text_font, text_size):
@@ -57,52 +71,72 @@ def text_to_video(text, voice, rate, pitch, video_width, video_height, bg_color,
         audio_clips.append(audio_clip)
 
         # 使用 wand 生成视频片段
-        with Image(width=video_width, height=video_height, background=Color(bg_color)) as img:
-            with Drawing() as draw:
-                draw.font = font_path
-                draw.font_size = text_size
-                draw.fill_color = Color(text_color)
-                draw.text_alignment = 'center'
-                draw.text_interline_spacing = 10
-                
-                # Calculate the y position for each line of text
+        with Drawing() as draw:
+            draw.font = font_path
+            draw.font_size = text_size
+            draw.fill_color = Color(text_color)
+            draw.text_alignment = 'center'
+            draw.text_interline_spacing = 10
+            
+            with Image(width=video_width, height=video_height, background=Color(bg_color)) as img:
                 lines = page.split("\n")
-                y_position = int((video_height - (len(lines) * (text_size + 10))) / 2)  # Center vertically
+                # Centering text vertically
+                total_text_height = len(lines) * (text_size + 10)  # Height of text area
+                start_y = (video_height - total_text_height) // 2  # Start position to center vertically
                 
-                for line in lines:
-                    draw.text(int(video_width / 2), y_position, line)
-                    y_position += int(text_size + 10)  # Move down for the next line (ensure y_position is an integer)
+                for j, line in enumerate(lines):
+                    draw.text(int(video_width / 2), start_y + (j * (text_size + 10)), line)
                 
-                draw(img)  # Draw the text on the image
+                draw(img)  # Apply the drawing to the image
+                img.format = 'png'
+                img_path = os.path.join(tempfile.gettempdir(), f"page_{i}.png")
+                img.save(filename=img_path)
+                text_clip = ImageClip(img_path).set_duration(audio_clip.duration).set_audio(audio_clip)
+                video_clips.append(text_clip)
 
-            img.format = 'png'
-            img_path = os.path.join(tempfile.gettempdir(), f"page_{i}.png")
-            img.save(filename=img_path)
-            text_clip = ImageClip(img_path).set_duration(audio_clip.duration).set_audio(audio_clip)
-            video_clips.append(text_clip)
-        
     # 合并所有视频片段
     final_video = concatenate_videoclips(video_clips)
     final_video_path = os.path.join(tempfile.gettempdir(), "output_video.mp4")
     final_video.write_videofile(final_video_path, fps=24, codec="libx264")
     return final_video_path, None
 
-# 示例函数调用（请根据需要调整）
-if __name__ == "__main__":
-    # 测试功能
-    text = "你好，这是一个测试视频。"
-    voice = "zh"
-    rate = 1.0
-    pitch = 1.0
-    video_width = 640
-    video_height = 480
-    bg_color = "white"
-    text_color = "black"
-    text_font = "path/to/your/font.ttf"  # 更新为实际字体路径
-    text_size = 30
+# Gradio接口函数
+def tts_interface(text, voice, rate, pitch, video_width, video_height, bg_color, text_color, text_font, text_size):
+    video, warning = text_to_video(text, voice, rate, pitch, video_width, video_height, bg_color, text_color, text_font, text_size)
+    return None, video, warning
 
-    video_path, warning = text_to_video(text, voice, rate, pitch, video_width, video_height, bg_color, text_color, text_font, text_size)
-    if warning:
-        print("Warning:", warning)
-    else:
-        print("Video saved at:", video_path)
+# 创建Gradio应用
+async def create_demo():
+    voices = await get_voices()
+    
+    demo = gr.Interface(
+        fn=tts_interface,
+        inputs=[
+            gr.Textbox(label="Input Text", lines=5),
+            gr.Dropdown(choices=[""] + list(voices.keys()), label="Select Voice", value=""),
+            gr.Slider(minimum=-50, maximum=50, value=0, label="Rate Adjustment (%)", step=1),
+            gr.Slider(minimum=-20, maximum=20, value=0, label="Pitch Adjustment (Hz)", step=1),
+            gr.Slider(minimum=640, maximum=1920, value=1080, label="Video Width", step=10),
+            gr.Slider(minimum=480, maximum=1080, value=720, label="Video Height", step=10),
+            gr.ColorPicker(value="#000000", label="Background Color"),
+            gr.ColorPicker(value="#FFFFFF", label="Text Color"),
+            gr.Textbox(label="Text Font", value="msyh.ttf"),  # 请确保字体文件路径正确
+            gr.Slider(minimum=10, maximum=100, value=24, label="Text Size", step=1)
+        ],
+        outputs=[
+            gr.Audio(label="Generated Audio", type="filepath"),
+            gr.Video(label="Generated Video"),
+            gr.Markdown(label="Warning", visible=False)
+        ],
+        title="Edge TTS Text to Speech and Video",
+        description="Convert text to speech and video using Microsoft Edge TTS. Adjust rate and pitch: 0 is the default value, positive values increase, and negative values decrease.",
+        analytics_enabled=False,
+        allow_flagging=False,
+    )
+    
+    return demo
+
+# 运行应用
+if __name__ == "__main__":
+    demo = asyncio.run(create_demo())
+    demo.launch()