TTS return empty wav file after training a model

58 views Asked by At

I have trained a tts model with LJSpeech-1.1 dataset, this is this config:

config = GlowTTSConfig(     batch_size=32,     eval_batch_size=16,     num_loader_workers=4,     num_eval_loader_workers=4,     run_eval=True,     test_delay_epochs=-1,     epochs=100,     text_cleaner="phoneme_cleaners",     use_phonemes=True,     phoneme_language="en-us",     phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),     print_step=25,     print_eval=False,     mixed_precision=True,     output_path=output_path,     datasets=[dataset_config],     save_step=1000,     )

After training my model i tried to generate a speech with the following command:

tts --text "Hello this is a test" --model_path '/tts_start/run-November-03-2023_08+59AM-0000000/checkpoint_1000.pth' --config_path '/tts_start/run-November-03-2023_08+59AM-0000000/config.json' --out_path 'out_2023_04_11.wav'

I got a wav file but no voice was generated, Does anyone know where i got wrong?

I followed instructions on this link https://colab.research.google.com/github/coqui-ai/TTS/blob/dev/notebooks/Tutorial_2_train_your_first_TTS_model.ipynb#scrollTo=d4c07f99-3d1d-4bea-801e-9f33bbff0e9f

0

There are 0 answers