update

.gitignore

@@ -6,6 +6,7 @@
 /dotenv/
 /logs/
 /trained_models
+/temp/
 
 **/__pycache__/
 

main.py

@@ -3,18 +3,21 @@
 import argparse
 import asyncio
 from functools import partial
+import json
 import logging
 from pathlib import Path
 import platform
 import re
 import tempfile
 from typing import List
+import uuid
 
-from project_settings import project_path, log_directory
+from project_settings import project_path, log_directory, edge_tts_record_file, edge_tts_temp_directory
 import log
 
 log.setup(log_directory=log_directory)
 
+import aiofiles
 import anyio
 import edge_tts
 import gradio as gr
@@ -47,11 +50,17 @@ async def edge_tts_get_speakers() -> List[str]:
 
 async def edge_tts_text_to_speech(text: str, speaker: str):
     communicate = edge_tts.Communicate(text, speaker)
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
-        tmp_path = tmp_file.name
-        await communicate.save(tmp_path)
 
-    return tmp_path
+    filename = edge_tts_temp_directory / "{}.wav".format(uuid.uuid4())
+    async with aiofiles.open(edge_tts_record_file.as_posix(), "a+", encoding="utf-8") as f:
+        await f.write(json.dumps({
+            "text": text,
+            "speaker": speaker,
+            "filename": filename.as_posix(),
+        }, ensure_ascii=False))
+
+    await communicate.save(filename)
+    return filename
 
 
 def shell(cmd: str):
@@ -71,11 +80,20 @@ def main():
         gr.Markdown(value=title)
 
         with gr.Tabs():
+            with gr.TabItem("Ebook Reading"):
+                e_book_reading_file = gr.File(
+                    value=(project_path / "data/e_book/confucianism/the_analects.txt").as_posix(),
+                    label="txt"
+                )
+                e_book_reading_tts_engine = gr.Dropdown(
+                    choices=["Edge TTS"], value="Edge TTS", label="tts_engine"
+                )
+                e_book_reading_button = gr.Button(value="e_book_reading", variant="primary")
             with gr.TabItem("Edge TTS"):
                 edge_tts_text = gr.Textbox(value="学而时习之，不亦悦乎。", lines=4, max_lines=50, label="text")
-                edge_tts_speaker = gr.Dropdown(choices=edge_tts_speakers_choices, value=edge_tts_speakers_choices[0], label="speakers")
+                edge_tts_speaker = gr.Dropdown(choices=edge_tts_speakers_choices, value="zh-CN-XiaoxiaoNeural", label="speakers")
 
-                edge_tts_audio = gr.Audio(type="filepath", label="audio")
+                edge_tts_audio = gr.Audio(type="filepath", label="audio", autoplay=True)
 
                 edge_tts_button = gr.Button(value="edge_tts", variant="primary")
             edge_tts_button.click(

project_settings.py

@@ -15,6 +15,11 @@ log_directory.mkdir(parents=True, exist_ok=True)
 temp_directory = project_path / "temp"
 temp_directory.mkdir(parents=True, exist_ok=True)
 
+edge_tts_temp_directory = temp_directory / "edge_tts"
+edge_tts_temp_directory.mkdir(parents=True, exist_ok=True)
+
+edge_tts_record_file = edge_tts_temp_directory / "edge_tts.jsonl"
+
 environment = EnvironmentManager(
     path=os.path.join(project_path, "dotenv"),
     env=os.environ.get("environment", "dev"),