返回的音频没有文件头，所以我随便加的一个文件头，但显示的音频时长有问题。如果播放器不能正常播放，可以使用Chrome播放。

xiaoxiao_tts.py

#!/usr/bin/env python
# -*- coding:utf-8 -*-
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
from ws4py.client.threadedclient import WebSocketClient
import binascii

class WSClient(WebSocketClient):
    def __init__(self, url, text, filename):
        self.fp = open(filename, 'wb')
        self.fp.write(binascii.unhexlify('524946460000000057415645666d74201000000001000200803e000000fa0000040010006461746100000000'))
        self.text = text
        super(WSClient, self).__init__(url)
        
    def opened(self):
        self.send('Content-Type:application/json; charset=utf-8\r\n\r\nPath:speech.config\r\n\r\n{"context":{"synthesis":{"audio":{"metadataoptions":{"sentenceBoundaryEnabled":"false","wordBoundaryEnabled":"true"},"outputFormat":"audio-24khz-48kbitrate-mono-mp3"}}}}\r\n')
        self.send("X-RequestId:fe83fbefb15c7739fe674d9f3e81d38f\r\nContent-Type:application/ssml+xml\r\nPath:ssml\r\n\r\n<speak version='1.0' xmlns='http://www.w3.org/2001/10/synthesis' xml:lang='en-US'><voice  name='Microsoft Server Speech Text to Speech Voice (zh-CN, XiaoxiaoNeural)'><prosody pitch='+0Hz' rate ='+0%' volume='+0%'>"+self.text+"</prosody></voice></speak>\r\n")

    def received_message(self, m):
        if 'turn.end' in m.data:
            self.close()
            self.fp.close()
        elif 'Path:audio\r\n' in m.data:
            self.fp.write(m.data.split('Path:audio\r\n')[1])
        else:
            # print(m)
            pass


if __name__ == '__main__':
    url = 'wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E9FB37E23D68491D6F4'
    text = '浙江温州，浙江温州，最大皮革厂，江南皮革厂倒闭了！老板黄鹤吃喝嫖赌，欠下了3.5个亿，带着他的小姨子跑了。我们没有办法，拿着钱包抵工资。原价都是三百多、二百多、一百多的钱包，通通二十块，通通二十块！黄鹤你不是人，我们辛辛苦苦给你干了大半年，你不发工资，你还我血汗钱，还我血汗钱！'
    filename = '/tmp/test.wav'
    ws = WSClient(url, text, filename)
    ws.connect()
    ws.run_forever()