在音频录制代码中添加一个计时器来记录实际录制时间，并将其写入OPUS音频流的首部。这将确保在播放OPUS文件时，播放器可以正确地计算时间戳，并正确地播放音频。

以下是一个示例代码段来实现这一方法：

import time
import sounddevice as sd
import soundfile as sf

# 设置录制参数
channels = 2
samplerate = 44100
blocksize = 2048
duration = 10  # seconds
filename = 'recording.opus'

# 记录实际录制时间
start_time = time.time()

# 记录OPUS音频流
with sf.SoundFile(filename, mode='w', samplerate=samplerate,
                  channels=channels, format='OGG', subtype='OPUS') as file:
    with sd.InputStream(samplerate=samplerate, blocksize=blocksize,
                         channels=channels) as stream:
        for i in range(int(duration * samplerate / blocksize)):
            block, overflow = stream.read(blocksize)
            file.write(block)

# 将记录的时间戳写入OPUS首部
with open(filename, 'rb+') as file:
    file.write(b'OpusHead')
    file.write(b'\x01\x00')
    file.write(b'\x00' * 7)
    timestamp = int((time.time() - start_time) * 48000)
    file.write(timestamp.to_bytes(8, byteorder='big'))

这段代码使用Python的sounddevice和soundfile库来记录OPUS音频流，并在录制结束时将实际时间戳写入OPUS流的首部。这将确保在播放OPUS文件时，播放器可以正确地计算时间戳，并正确地播放音频。