Fastspeech2 training error : Not found any mels files in $./dump_ljspeech/train/.

[neso613]

Hi all, I have just started learning TTS model building. I am trying to train fastspeech2 model using ljspeech dataset. For that, I have donwloaded ljspeech dataset and run folllowing commands - tensorflow-tts-preprocess --rootdir ./ljspeech --outdir ./dump_ljspeech --config preprocess/ljspeech_preprocess.yaml --dataset ljspeech

tensorflow-tts-normalize --rootdir ./ljspeech --outdir ./dump_ljspeech --config preprocess/ljspeech_preprocess.yaml --dataset ljspeech

norm-feats folder created but have no items in it.

Following commands used to start training - `CUDA_VISIBLE_DEVICES=0 python examples/fastspeech2/train_fastspeech2.py \

--train-dir ./dump_ljspeech/train/ \ --dev-dir ./dump_ljspeech/valid/ \ --outdir ./examples/fastspeech2/exp/train.fastspeech2.v1/ \ --config ./examples/fastspeech2/conf/fastspeech2.v1.yaml \ --use-norm 1 \ --f0-stat ./dump_ljspeech/stats_f0.npy \ --energy-stat ./dump_ljspeech/stats_energy.npy \ --mixed_precision 1 \ --resume "" ` But end up with this error-

Why training is failing? @dathudeptrai @ZDisket

[neso613]

looking forward to hearing from the team and community.

[neso613]

@dathudeptrai can you pls share your thoughts on this.

[neso613]

@machineko @erogol pls help me in this regard.

[mmseerrttt]

@neso613 I also encountered the same problem, but with tacotron2. have you solved it.

[mmseerrttt]

you can try

tensorflow-tts-preprocess --rootdir ./ljspeech --outdir ./dump_ljspeech --config preprocess/ljspeech_preprocess.yaml --dataset ljspeech
tensorflow-tts-normalize --rootdir ./dump_ljspeech --outdir ./dump_ljspeech --config preprocess/ljspeech_preprocess.yaml --dataset ljspeech

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.