Update app.py

app.py

@@ -4,9 +4,6 @@ import asyncio
 import tempfile
 import os
 from moviepy.editor import AudioFileClip
-from wand.image import Image
-from wand.drawing import Drawing
-from wand.color import Color
 
 # 获取所有可用的语音
 async def get_voices():
@@ -30,18 +27,21 @@ async def text_to_speech(text, voice, rate, pitch):
     return tmp_path, None
 
 # SRT文件生成
-def generate_srt(pages, audio_clips):
+def generate_srt(words, audio_clips):
     srt_path = os.path.join(tempfile.gettempdir(), "output_subtitles.srt")
     with open(srt_path, 'w', encoding='utf-8') as srt_file:
-        for i, (page, audio_clip) in enumerate(zip(pages, audio_clips)):
-            start_time = sum(audio_clip.duration for audio_clip in audio_clips[:i])
+        current_time = 0
+        for i, (word, audio_clip) in enumerate(zip(words, audio_clips)):
+            start_time = current_time
             end_time = start_time + audio_clip.duration
             
             # Convert to SRT format
             start_time_str = format_srt_time(start_time)
             end_time_str = format_srt_time(end_time)
-            srt_file.write(f"{i + 1}\n{start_time_str} --> {end_time_str}\n{page}\n\n")
-    
+            srt_file.write(f"{i + 1}\n{start_time_str} --> {end_time_str}\n{word}\n\n")
+            
+            current_time += audio_clip.duration  # Update current time
+
     return srt_path
 
 def format_srt_time(seconds):
@@ -55,42 +55,24 @@ def format_srt_time(seconds):
 
 # 文字转音频和SRT功能
 async def text_to_audio_and_srt(text, voice, rate, pitch):
-    # 计算每页可以容纳的行数和每行可以容纳的字符数
-    max_chars_per_line = 60  # 适当设置每行最大字符数
-    max_lines_per_page = 5    # 每页最大行数
-
-    # 按页拆分文本
+    # Split text into words
     words = text.split()
-    lines = []
-    current_line = ""
-    pages = []
-    
-    for word in words:
-        if len(current_line) + len(word) + 1 > max_chars_per_line:
-            lines.append(current_line)
-            current_line = word
-            if len(lines) == max_lines_per_page:
-                pages.append("\n".join(lines))
-                lines = []
-        else:
-            current_line = f"{current_line} {word}".strip()
-    
-    lines.append(current_line)
-    if lines:
-        pages.append("\n".join(lines))
-
-    # 为每页生成独立音频
     audio_clips = []
-    for page in pages:
-        audio_text = page.replace("\n", " ")  # 移除换行符以防止 TTS 停顿
-        audio, warning = await text_to_speech(audio_text, voice, rate, pitch)
+    subtitle_chunks = []
+
+    # Generate audio for each 3-4 word chunk
+    for i in range(0, len(words), 3):  # Change the step to 3 or 4 based on your preference
+        chunk = ' '.join(words[i:i + 3])  # Create a chunk of 3 words
+        audio, warning = await text_to_speech(chunk, voice, rate, pitch)
         if warning:
             return None, None, warning
+        
         audio_clip = AudioFileClip(audio)
         audio_clips.append(audio_clip)
+        subtitle_chunks.append(chunk)
 
     # 生成SRT文件
-    srt_path = generate_srt(pages, audio_clips)
+    srt_path = generate_srt(subtitle_chunks, audio_clips)
     return audio, srt_path, None
 
 # Gradio接口函数