From b225e8afd5013fef3542caf4dbbe91953367e209 Mon Sep 17 00:00:00 2001 From: hrz <1710360675@qq.com> Date: Wed, 28 May 2025 16:19:21 +0800 Subject: [PATCH] =?UTF-8?q?update:=E4=BC=98=E5=8C=96=E7=AC=AC=E4=B8=80?= =?UTF-8?q?=E5=8F=A5=E8=AF=9D?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../providers/tts/huoshan_double_stream.py | 25 ++++++++++++++++--- 1 file changed, 22 insertions(+), 3 deletions(-) diff --git a/main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py b/main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py index 1e76f538..2d635207 100644 --- a/main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py +++ b/main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py @@ -237,9 +237,12 @@ class TTSProvider(TTSProviderBase): async def _start_monitor_tts_response(self): opus_datas_cache = [] + # 添加标志来区分是否是第一句话 + is_first_sentence = True while not self.conn.stop_event.is_set(): try: - msg = await self.ws.recv() # 确保 `recv()` 运行在同一个 event loop + # 确保 `recv()` 运行在同一个 event loop + msg = await self.ws.recv() res = self.parser_response(msg) self.print_response(res, "send_text res:") @@ -258,10 +261,23 @@ class TTSProvider(TTSProviderBase): logger.bind(tag=TAG).debug( f"推送数据到队列里面帧数～～{len(opus_datas)}" ) - opus_datas_cache = opus_datas_cache + opus_datas + if is_first_sentence: + # 第一句话直接发送 + self.tts_audio_queue.put( + (SentenceType.MIDDLE, opus_datas, self.tts_text) + ) + else: + # 后续句子缓存 + opus_datas_cache = opus_datas_cache + opus_datas elif res.optional.event == EVENT_TTSSentenceEnd: logger.bind(tag=TAG).info(f"句子语音生成成功：{self.tts_text}") - self.tts_audio_queue.put((SentenceType.MIDDLE, opus_datas_cache, self.tts_text)) + if not is_first_sentence: + # 只有非第一句话才发送缓存的数据 + self.tts_audio_queue.put( + (SentenceType.MIDDLE, opus_datas_cache, self.tts_text) + ) + # 第一句话结束后，将标志设置为False + is_first_sentence = False elif res.optional.event == EVENT_SessionFinished: logger.bind(tag=TAG).debug(f"会话结束～～") for tts_file, text in self.before_stop_play_files: @@ -272,6 +288,9 @@ class TTSProvider(TTSProviderBase): ) self.before_stop_play_files.clear() self.tts_audio_queue.put((SentenceType.LAST, [], None)) + + opus_datas_cache = [] + is_first_sentence = True continue except websockets.ConnectionClosed: break # 连接关闭时退出监听