search

AlizerUncaged / desktop-waifu

Desktop Waifu!
334 stars 49 forks source link

There is no audio output for a reply. #207

Closed sato-kazuma-kun closed 8 months ago

sato-kazuma-kun commented 10 months ago

I'm no expert in this matter but I think something might be wrong with the generate_audio_bytes function in file- /waifu/Src/utils/elevenlabs.py line no. 32- -> mp3_bytes = ELEVENLABS_VOICE.generate_audio_bytes(message, stability, similarity)

This is the error I get everytime after reply message or the reply from AI- 'ElevenLabsVoice' object has no attribute 'generate_audio_bytes'. Screenshot 2023-10-24 145542

☝️ Here is an image for referance, everything works except there is no audio reply in the reply,

sato-kazuma-kun commented 10 months ago

So, I just found the code that works-

Just replace the line that was specified with this mp3_bytes = ELEVENLABS_VOICE.generate_audio_v2(message, GenerationOptions(model_id='eleven_multilingual_v1', stability=stability, similarity_boost=similarity))[0]

And it should work fine, atleast it works for me.