saiful islam
saiful islam
I have converted a Bert variant model which onnx file more faster then pytorch model ORT into geforce rtx 3090. Below the result for 3090. but when I used Tesla...
@bensonbs Thank you for your clear explanation, Could you please share some samples after applying DPO and the audio quality?
@harshkaria colab default python 3.10. This requirement files some module not support python 3.10. Degrade python version to 3.8 and install requirements file i think your issue will be solved.
I have the same question as like @BernardoOlisan of ```SpeechFeatureEmbedding``` class defined ```self.pos_emb = layers.Embedding(input_dim=maxlen, output_dim=num_hid)``` but not used. I need clarification also.