You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As given in the Alex Graves CTC paper, we sum over probabilities of all the possible alignments using dynamic programming which gives us certain transcription. In code and in tensorflow documentation it is not explicitly mentioned that where this part is happening. Could you help me here and tell me in which of the following functions, this DP thing is happening:
tf.nn.ctc_loss
tf.nn.ctc_beam_search_decoder
tf.nn.ctc_greedy_decoder
The text was updated successfully, but these errors were encountered:
As given in the Alex Graves CTC paper, we sum over probabilities of all the possible alignments using dynamic programming which gives us certain transcription. In code and in tensorflow documentation it is not explicitly mentioned that where this part is happening. Could you help me here and tell me in which of the following functions, this DP thing is happening:
The text was updated successfully, but these errors were encountered: