Murugan Rajenthiran

solved the issue. the problem is didn't get the right output feature map shape for **_activation_49_** ``` Traceback (most recent call last): File "cub_demo.py", line 102, in share_fea_map = Reshape((final_dim,...

# ValueError: Shape must be rank 2 but is rank 3 for 'MatMul_29' (op: 'MatMul') with input shapes: [?,196,512], [?,512]. ``` Traceback (most recent call last): File "/home/dell/Videos/Object_Tracking/env_a3mv2/lib/python3.6/site-packages/tensorflow/python/framework/common_shapes.py", line 671,...

Thank you so much for your reply sir. I want to build model for Speech to text. So if YouTube-8m dataset how to handle on YouTube audio segments with corresponding...

@PratikNalage cnn_daily_load.py you too create function like this, ``` def cnn_daily_load(): filenames=load_data(datasets["cnn"],data_categories[0]) """----------load the data, sentences and summaries-----------""" for k in range(len(filenames[:400])): if k%2==0: try: data["articles"].append(cleantext(parsetext(datasets["cnn"],data_categories[0],"%s"%filenames[k]))) except Exception as e:...

not really @DeepsMoseli. in this place you are using gensim - skipgram algorithm(word2vec) to build normal word2vec model then generating embedding for the words. training from scratch. great stuff. we...

@Edresson sir, Thank you so much for your excellent work. very nice paper. Multi-lingual Zero-shot pre-trained model, I tried with speaker adaptation (359 samples - 25 minutes). but it's not...

@WeberJulian Thank you so much. Do i need touch speaker Encoder model to Finetune with 1000 samples to get better reference speaker quality? Thanks

anyone know this?

I'm also same confusion. someone explain how to execute ?