Comments (7)
The problem here is in the path given to the enroll_stat
object. Could you post the conf.yaml
file as it contains all the paths and configurations that the project needs??
from speaker-recognition.
#---! THIS FILE SHOULD BE MODIFIED BASED ON YOUR CASE
#These variables should be changed
inpath: D:/Merged_Arabic_Corpus_of_Isolated_Words/Merged_Arabic_Corpus_of_Isolated_Words #where input wave data exist
outpath: D:/Merged_Arabic_Corpus_of_Isolated_Words/Allah #where the program outputs exist
#training configuration
#NOTE: the summation of following sessions must be less than or equal 10
enroll_sessions: 3 #number of sessions to be included in the training set
test_sessions: 5 #number of sessions to be included in the test set
#preprocessing configurations
sampling_rate: 44100
bit_precision: 16
no_channels: 1
#features configuration (NOTE: to disable a feature, assign it to None)
features:
- vad #voice-activity-dectection
- energy #log-energy
- cep #cepstral coefficients
- fb #filter-banks
#- bnf
cepstral_coefficients: 19
filter_bank: log #filter bank can either be "log" for logarithmic and "lin" for linear
filter_bank_size: 24
lower_frequency: 300
higher_frequency: 3400
vad: snr #can be either "energy", "snr", "percentil" or "lbl".
snr_ratio: 40 #signal-to-noise ratio
window_size: 0.025 #in seconds (25 milliseconds)
window_shift: 0.010 #in seconds (10 milliseconds)
#UBM configuration
num_gaussians: 64
#i-vectors (tv) configurations (depends on UBM configurations)
batch_size: 30 #size of data batch
tv_rank: 25 # Rank of the total variability matrix
tv_iterations: 50 # number of iterations to train the variability matrix
scoring: cosine #can be either "cosine", or "mahalanobis", or "two_covariance"
#i-vectors (PLDA) configurations (depends on TV configuration)
#plotting configuration
DET_curve: rocch #you can choose either 'rocch' or 'steppy'
#other configuration
from speaker-recognition.
Now, in the directory D:/Merged_Arabic_Corpus_of_Isolated_Words/Allah
... make sure of these things:
- This directory has these folders which are
audio
,feat
andtask
. - Then, make sure the
audio
folder has three folders inside itdata
,enroll
andtest
. - And make sure the
feat
directory has two folders inside itenroll
andtest
.
If that is the case, then please tell me how many files are in feat\enroll
and feat\test
?? You can do that simply by selecting all the files within the folder and Windows will tell you how many items are selected
from speaker-recognition.
feat\enroll 300 items
feat\test 500 items
from speaker-recognition.
As I understood from your mail, this isn't a problem anymore !!
from speaker-recognition.
I will close this for now but feel free to re-open it
from speaker-recognition.
I now have the same problem, but how to solve it?
from speaker-recognition.
Related Issues (14)
- Feature Server HOT 3
- HDF5 Exception HOT 2
- Accuracy metrics
- Not found enroll_idmap.h5 HOT 1
- run dada_init.py
- Using Kaldi model
- how to train ivector? HOT 2
- createFeatureServer() method is ambiguous HOT 2
- ValueError: need at least one array to concatenate HOT 5
- How to add a new speaker without retraining HOT 2
- second issue
- third issue
- Test on a new speaker HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from speaker-recognition.