24-hour audio using embedding model nnfp

[ToTheMonn]

Hi , I have a 24-hour audio , using NNFP for processing , reporting the error of not having enough memory . How can I fix the problem ? Here is the code.

emb_pipe = (
pipe.input('url')
.map('url', 'frames', ops.audio_decode.ffmpeg())
.map('frames', 'emb', ops.audio_embedding.nnfp())
)

[jaelgu]

@ToTheMonn How long is your audio each time running the pipe?