-
Attaching my loss card from tensorboard/ does everything looks okay? val_epoch seems to have converged, what were the final losses you were getting. Any other thing I need to keep an eye on?
![ima…
-
In the white paper, they mention conditioning to a particular speaker as an input they condition globally, and the TTS component as an up-sampled (deconvolution) conditioned locally. For the latter, t…
-
Thanks for your code, but I want to learn how to use your modle to train a new set of data, so can you provide a train.py file?
-
I have a trained voice pt file. when generating voice should i provide original training dataset sound too? does it matter or improve?
It is working very well
By the way i am using this branch
…
-
微博内容精选
-
Trying to get the lumina_music demo working.
I have all the models downloaded locally.
Using the command...
`python demo_music.py --ckpt ..\models\Lumina-Music\music_generation --vocoder_ckpt ..\m…
-
Have trained `update_v2` branch on :
* Extracted Semantic token from HuBert Large layer 16 with 1024 cluster Kmean. (`50 tok/sec`)
* Extracted Acoustic token from Encodec 24 khz sample rate, 240 ho…
-
Hello,
Thank you for sharing this dataset.
Would it be possible to have more information on the generation of audios? In particular, the names of the vocoders training datasets used.
Thank yo…
-
-
># Spectrogram VQGAN as a Neural Audio Codec
>model_name = '2021-05-19T22-16-54_vggsound_codebook'
>log_dir = './logs'
>config, model, vocoder = load_neural_audio_codec(model_name, log_dir, device)…