From b267b3230aa2d8ba4eb1369c82c6185906b1f4a1 Mon Sep 17 00:00:00 2001
From: hrz <1710360675@qq.com>
Date: Tue, 11 Feb 2025 13:51:44 +0800
Subject: [PATCH] =?UTF-8?q?update:=E4=BC=98=E5=8C=96stt=E5=8F=91=E9=80=81?=
 =?UTF-8?q?=E9=A1=BA=E5=BA=8F?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 core/handle/audioHandle.py | 45 +++++++++++++++++++-------------------
 1 file changed, 23 insertions(+), 22 deletions(-)

diff --git a/core/handle/audioHandle.py b/core/handle/audioHandle.py
index 4e55d115..7217178c 100644
--- a/core/handle/audioHandle.py
+++ b/core/handle/audioHandle.py
@@ -36,19 +36,10 @@ async def handleAudioMessage(conn, audio):
         conn.reset_vad_states()
 
 async def startToChat(conn, text):
-    stt_text = get_string_no_punctuation_or_emoji(text)
-    await conn.websocket.send(json.dumps({
-        "type": "stt",
-        "text": stt_text,
-        "session_id": conn.session_id}
-    ))
-    await conn.websocket.send(
-        json.dumps({
-            "type": "llm",
-            "text": "😊",
-            "emotion": "happy",
-            "session_id": conn.session_id}
-        ))
+    # 异步发送 stt 信息
+    asyncio.create_task(
+        schedule_with_interrupt(0, send_stt_message(conn, text))
+    )
     conn.executor.submit(conn.chat, text)
 
 
@@ -59,10 +50,6 @@ async def sendAudioMessage(conn, audios, duration, text):
     if text == conn.tts_first_text:
         logger.info(f"发送第一段语音: {text}")
         conn.tts_start_speak_time = time.time()
-        await send_tts_message(conn, "start", text)
-
-    # 先等待0.5秒，确保客户端处于 speaking 状态
-    await asyncio.sleep(0.5)
 
     # 发送 sentence_start（每个音频文件之前发送一次）
     await send_tts_message(conn, "sentence_start", text)
@@ -78,15 +65,12 @@ async def sendAudioMessage(conn, audios, duration, text):
             await send_tts_message(conn, "sentence_end", text)
 
     if conn.llm_finish_task and text == conn.tts_last_text:
-        stop_duration = conn.tts_duration - \
-            (time.time() - conn.tts_start_speak_time)
+        stop_duration = conn.tts_duration - (time.time() - conn.tts_start_speak_time)
         stop_task = asyncio.create_task(
-            schedule_with_interrupt(
-                stop_duration, send_tts_message(conn,  'stop'))
+            schedule_with_interrupt(stop_duration, send_tts_message(conn, 'stop'))
         )
         conn.scheduled_tasks.append(stop_task)
 
-
 async def send_tts_message(conn, state, text=None):
     """发送 TTS 状态消息"""
     message = {
@@ -101,6 +85,23 @@ async def send_tts_message(conn, state, text=None):
     if state == "stop":
         conn.clearSpeakStatus()
 
+async def send_stt_message(conn, text):
+    """发送 STT 状态消息"""
+    stt_text = get_string_no_punctuation_or_emoji(text)
+    await conn.websocket.send(json.dumps({
+        "type": "stt",
+        "text": stt_text,
+        "session_id": conn.session_id}
+    ))
+    await conn.websocket.send(
+        json.dumps({
+            "type": "llm",
+            "text": "😊",
+            "emotion": "happy",
+            "session_id": conn.session_id}
+        ))
+    await send_tts_message(conn, "start")
+
 
 async def schedule_with_interrupt(delay, coro):
     """可中断的延迟调度"""