diff --git a/main/xiaozhi-server/core/utils/util.py b/main/xiaozhi-server/core/utils/util.py
index 641e2655..088d8962 100644
--- a/main/xiaozhi-server/core/utils/util.py
+++ b/main/xiaozhi-server/core/utils/util.py
@@ -5,6 +5,8 @@ import re
 import os
 import wave
 from io import BytesIO
+from typing import Callable, Any
+
 from core.utils import p3
 import numpy as np
 import requests
@@ -245,6 +247,23 @@ def audio_bytes_to_data(audio_bytes, file_type, is_opus=True):
         raw_data = audio.raw_data
         return pcm_to_data(raw_data, is_opus)
 
+def audio_bytes_to_data_stream(audio_bytes, file_type, is_opus, callback: Callable[[Any], Any]):
+    """
+    直接用音频二进制数据转为opus/pcm数据，支持wav、mp3、p3
+    """
+    if file_type == "p3":
+        # 直接用p3解码
+        return p3.decode_opus_from_bytes_stream(audio_bytes, callback)
+    else:
+        # 其他格式用pydub
+        audio = AudioSegment.from_file(
+            BytesIO(audio_bytes), format=file_type, parameters=["-nostdin"]
+        )
+        audio = audio.set_channels(1).set_frame_rate(16000).set_sample_width(2)
+        raw_data = audio.raw_data
+        pcm_to_data_stream(raw_data, is_opus, callback)
+        return None
+
 
 def pcm_to_data(raw_data, is_opus=True):
     # 初始化Opus编码器
@@ -276,6 +295,32 @@ def pcm_to_data(raw_data, is_opus=True):
 
     return datas
 
+def pcm_to_data_stream(raw_data, is_opus=True, callback: Callable[[Any], Any]=None):
+    # 初始化Opus编码器
+    encoder = opuslib_next.Encoder(16000, 1, opuslib_next.APPLICATION_AUDIO)
+
+    # 编码参数
+    frame_duration = 60  # 60ms per frame
+    frame_size = int(16000 * frame_duration / 1000)  # 960 samples/frame
+
+    # 按帧处理所有音频数据（包括最后一帧可能补零）
+    for i in range(0, len(raw_data), frame_size * 2):  # 16bit=2bytes/sample
+        # 获取当前帧的二进制数据
+        chunk = raw_data[i : i + frame_size * 2]
+
+        # 如果最后一帧不足，补零
+        if len(chunk) < frame_size * 2:
+            chunk += b"\x00" * (frame_size * 2 - len(chunk))
+
+        if is_opus:
+            # 转换为numpy数组处理
+            np_frame = np.frombuffer(chunk, dtype=np.int16)
+            # 编码Opus数据
+            frame_data = encoder.encode(np_frame.tobytes(), frame_size)
+            callback(frame_data)
+        else:
+            frame_data = chunk if isinstance(chunk, bytes) else bytes(chunk)
+            callback(frame_data)
 
 def opus_datas_to_wav_bytes(opus_datas, sample_rate=16000, channels=1):
     """