Closed simdax closed 9 months ago
Hi @simdax, the Effnet model you are using requires a minimum audio length of ~60 seconds. In standard mode we simply pad shorter audios, but the feature is not implemented in streaming mode.
Could you try with a longer audio to verify your implementation?
Reopen if this is still an issue.
Hello all
I have to admit I'm not very familiar with AI, so I'm struggling to create a simple example that would work as the other tensorflow examples, musicnn or vggish, in CPP.
In python I have a result with this code:
I've just copied these files which are all similar, but it does not seem to work for me sadly.
Anyone could help :) ? Thank you
compiling it and activating it like these
./build/src/examples/essentia_streaming_discogs test/models/effnetdiscogs/effnetdiscogs-bs64-1.pb test/audio/recorded/mozart_c_major_30sec.wav outpout.json
the result is empty :(
Thank you very much :)