litellm - 💡(How to fix) Fix 视频抽帧分析 [1 participants]

2026-04-03 14:03:07 | ERROR | "./app\services\voice.py:1180": azure_tts_v1 - 生成音频文件时出错: 403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E9FB37E23D68491D6F4&Sec-MS-GEC=848195204AF65BBC6781ED56BBC9AEEBAABCCA959062FF4DDCEA940E2BFD4C99&Sec-MS-GEC-Version=1-130.0.2849.68&ConnectionId=6e9db8d439f944d38f5bcb4298ad5d04'

TL;DR

The issue is likely due to an authentication or authorization problem with the Azure TTS service, and checking the TrustedClientToken and Sec-MS-GEC parameters may help resolve the 403 error.

Guidance

• Verify that the TrustedClientToken and Sec-MS-GEC parameters are correctly generated and valid for the request.
• Check the Azure TTS service documentation to ensure that the authentication and authorization headers are properly formatted and included in the request.
• Inspect the ConnectionId parameter to ensure it is correctly generated and matches the expected format.
• Review the error message and response status code to determine if there are any specific requirements or restrictions for the request.

Notes

The issue seems to be related to authentication or authorization with the Azure TTS service, but without more information about the request and the service configuration, it is difficult to provide a more specific solution.

Recommendation

Apply workaround: Verify and update the authentication and authorization parameters to ensure they are correct and valid for the request, as the 403 error suggests an issue with the request authentication.