-
Hi, thank you for you nice work. I have successfully run the project, however I noticed that my gpu resources remained unused and the attacking step (running attackMain.sh) has taken me more than 18 h…
-
Does this repo works with verification i.e the speaker not present in dataset is predicted as invalid voice i need to integrate this for login and registration with voice
please reply
thanks
-
Hi Daniel,
Dev env: ARM 64.
What to do: extract ivectors
the command:
`sid/extract_ivectors.sh --cmd run.pl --num-threads 4 --nj 1 model/extractor_1024 xxx/data xxx/exp/ivector_1024
sid/extr…
-
Hi there,
Thanks for sharing the code. I wonder if you can help on the issue described below.
I am following the instruction step and step to try the code. I used the pre-models you pr…
-
The `get_means()`, `get_vars()` and `weights_` for GMM classes return `_kaldi_matrix.Matrix` and `_kaldi_vector.Vector` instances rather than Matrix or Vector instances. These instances don't support …
-
- [x] Construir UBM:
- [x] Extrar MFCCs de los audios de los spekares usando 20 coeficientes, normalizando el audio antes:
- [x] 225
- [x] 228
- [x] 229
- [x] 230
- [x] 231
- [x] 233
- [x] 2…
-
```
# gmm-global-get-post --n=5 --min-post=0.025 extractor/final.dubm "ark,s,cs:apply-cmvn-online --spk2utt=ark:data/train/split1/1/split1/1/spk2utt --config=extractor/online_cmvn.conf extractor/glob…
-
Hi,
Thank you for providing a nice repository. I have a question regarding training a gmm-ubm model.
I have generated MFCC vectors for my data (binary classification task. Data is in the form of…
-
Hi everyone,
Recently I read about the training process of UBM, but I have some confusion in the process in kaldi in file mle-diag-gmm.cc, line 325 and 333,
I don't know why `mean.Scale(1.0 / occ…
-
Managed to do mfcc and vad - checked data to be valid.
Now when train_diag_ubm is ran I get error job failed without further information
Utterances are relative long, it's 90 people each recorded fo…