Add TTS markdown sanitization and expand deployment docs.

Strip Markdown and stage directions before ChatTTS synthesis with chunked long scripts; document model pre-download, server-update, and microphone HTTPS notes.

Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
dekun
2026-06-12 16:31:06 +08:00
parent 0f5277c22e
commit f36056d293
6 changed files with 409 additions and 50 deletions
+8
View File
@@ -74,6 +74,11 @@ SYSTEM_PROMPT = (
"如果原视频中有由于心态不好、违背交易纪律(如手贱乱开仓、提前平仓)"
"导致少赚或亏损的部分,请用冷酷、严厉的语气狠狠地自我吐槽、反思该点。"
"去掉所有无意义的口头禅,字数不做删减。"
"【输出格式硬性要求】"
"只输出可直接朗读的纯文本正文,不要 Markdown(禁止 #、**、---、列表符号、emoji)。"
"不要写舞台提示(如前奏、转场、BGM、语气说明等括号备注)。"
"不要写「以下是润色后的文案」等前言,也不要写修改笔记或点评。"
"可用《》作为标题,正文按自然段换行即可。"
)
# ---------------------------------------------------------------------------
@@ -131,6 +136,9 @@ TTS_TOP_P = 0.7
TTS_TOP_K = 20
TTS_SPEED_PROMPT = "[speed_5]"
# 单段 TTS 最大字数(超长稿按句切分后逐段合成再拼接)
TTS_MAX_CHARS_PER_CHUNK = _env_int("TTS_MAX_CHARS_PER_CHUNK", 280)
# ---------------------------------------------------------------------------
# 上传临时文件目录
# ---------------------------------------------------------------------------