AudioLM Generation Script

[cuongnguyengit]

Hi, thanks for your power project.
I have a question when the audio lm model inference for TTS task.
About TTS, we need to put the text like an input of the TTS model and i can't find any code to do it in the audilom-pytorch source.
I think that AudioLM only changed text embedding (of any TTS model) to LLM (like T5).

[LWprogramming]

Hi! This is just my training scripts and I'd check out the actual repo which these scripts are written for, e.g. lucidrains/audiolm-pytorch#32, for some ideas on including text input.