Update app.py

app.py

@@ -4,9 +4,6 @@ import asyncio
 import tempfile
 import os
 from moviepy.editor import AudioFileClip, concatenate_audioclips
-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():
@@ -30,20 +27,23 @@ async def text_to_speech(text, voice, rate, pitch):
     return tmp_path, None
 
 # 生成SRT文件
-def generate_srt(words, audio_clips):
+def generate_srt(words, audio_duration, fps=24):
     srt_path = os.path.join(tempfile.gettempdir(), "output_subtitles.srt")
     with open(srt_path, 'w', encoding='utf-8') as srt_file:
+        # Split audio duration into segments for SRT
+        segment_duration = audio_duration / len(words)  # Average duration per word
         current_time = 0
-        for i, (word, audio_clip) in enumerate(zip(words, audio_clips)):
+        
+        for i, word in enumerate(words):
             start_time = current_time
-            end_time = start_time + audio_clip.duration
+            end_time = start_time + segment_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{word}\n\n")
             
-            current_time += audio_clip.duration  # Update current time
+            current_time += segment_duration  # Update current time
 
     return srt_path
 
@@ -58,30 +58,19 @@ 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):
-    # Split text into words
-    words = text.split()
-    audio_clips = []
-    subtitle_chunks = []
-
-    # Generate audio for each chunk of 3 words
-    for i in range(0, len(words), 3):  # Adjust chunk size if needed
-        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)
+    # Generate audio for the entire text
+    audio_path, warning = await text_to_speech(text, voice, rate, pitch)
+    if warning:
+        return None, None, warning
 
-    # Concatenate all audio clips
-    final_audio_path = os.path.join(tempfile.gettempdir(), "output_audio.mp3")
-    final_audio = concatenate_audioclips(audio_clips)
-    final_audio.write_audiofile(final_audio_path)
+    audio_clip = AudioFileClip(audio_path)
+    audio_duration = audio_clip.duration
+    
+    # Generate SRT file based on the entire text
+    words = text.split()
+    srt_path = generate_srt(words, audio_duration)
 
-    # 生成SRT文件
-    srt_path = generate_srt(subtitle_chunks, audio_clips)
-    return final_audio_path, srt_path, None
+    return audio_path, srt_path, None
 
 # Gradio接口函数
 def tts_interface(text, voice, rate, pitch):