#7792 Each line must contain 3 parts separated by '#': audio_name.wav#audio_text#lang_code

The reference-audio line format you wrote (audio_name.wav#audio_text#lang_code) is not the format pyVideoTrans accepts.

The correct format has only 2 parts, separated by a single #:

filename.wav#the spoken text in that audio

Steps:

1. Prepare a mono WAV, 3–10 seconds, single clean voice, no background noise, no leading/trailing silence.
2. Copy the file into the f5-tts folder inside the software directory (yours: C:/Users/1234/Desktop/Pyvideotrans/f5-tts).
3. Open Menu -> TTS Settings -> Set Reference Audio and enter: myaudio1.wav#the exact words spoken in that audio
4. Save, then select myaudio1.wav in the Voice Role dropdown on the main screen.

Notes:

• Do not append a language code; language comes from the Main UI Target Language / TTS channel settings, not from the reference line.
• For GPT-SoVITS the reference wav must be placed in the GPT-SoVITS root folder, not f5-tts.
• If you want cloning to use the original video's voice instead, choose the clone voice role (video translation mode only).

请查阅相关文档：

• 语音克隆与多角色配音
• 全部配音渠道说明