sanakhamekhem

Error installing torch 7

2

Building on 8 cores -- The C compiler identification is GNU 4.8.5 -- The CXX compiler identification is GNU 4.8.5 -- Check for working C compiler: /usr/bin/cc -- Check for...

Confidence score computing

1

I'm using the following code to get the transcription: beam_result, beam_scores, timesteps, out_seq_len = decoder.decode(probs_seq) Now, I would like t get the confidence score from the beam_scores variable. Is it...

I'm using the ctcdecode implementation in https://github.com/SeanNaren/ctcdecode. I would like to decode without lexicon but using a language model at character level. Is that possible??

I'm asking if I can use a word lexicon at the decoding. Thank you in advance.

Dear Sir, I'm planning to use this code to extract named entities from frensh texts. The tags are : wife name, husband name, mother name, father name, birth day, etc....

word Lattice to phone lattice / to slf

4

Hi Alls, I would like to convert a word lattice generated using eesen to phone level lattice like in kaldi (lattice-to-phone-lattice 1.mdl ark:1.lats ark:phones.lats). Is there a sollution? Is it...

Hi, I would like to get the feature vector extracted using the CNN, how can I get from the last conv layer in this source code. Thank you in advance

Dear Sir, I'm intersted to this code. I'm building a recognizer for images having a height of 128 pixels. How can I update just the CNN recognier architecture's model to...

### I'm building a text recogniser using eesen framework, the training step is realised succesfully. But the decoding stage fails. I have changed the beam,lattice-beam and acoustic scale many times....

Hi, Thanks for this great work. I will be greatfull if you can provide me by the pretrained weights model's docdiff that were used for the H-Dibco datasets for research...