Stack Overflow
Stack Overflow

Questions tagged [kaldi]

Kaldi speech recognition toolkit

Kaldi is a toolkit for speech recognition written in C++ and licensed under the Apache License v2.0. Kaldi is intended for use by speech recognition researchers.

113 questions
0
votes
1 answer

How to convert a Python dictionary to a Numpy array?

So the logistic regression from the sklearn library from Python has the .fit() function which takes x_train(features) and y_train(labels) as arguments to train the classifier. It seems that x_train.shape = (number_of_samples, number_of_features) For…
Petar Yakov
  • 169
  • 2
  • 14
0
votes
1 answer

Error constructing TableReader: rspecifier is scp:data/train/split1/1/feats.scp

I followed the kaldi for dummies tutorial but I encounter this error on the MONO TRAINING part: ===== MONO TRAINING ===== steps/train_mono.sh --nj 1 --cmd run.pl data/train data/lang exp/mono steps/train_mono.sh: Initializing monophone…
chip
  • 3,039
  • 5
  • 35
  • 59
0
votes
1 answer

Speech to text in dot net core 3.1 web app without using cloud services like azure

I am trying to implement speech to text on my .net core 3.1 web app without using any cloud services like Azure, AWS, or Google. I have found options like deepspeech, kaldi,pocketpheonix. It looks like they can be implemented on a .net core app but…
Abid
  • 5
  • 1
0
votes
1 answer

weird awk outputs in reading/writing file

I'm working on a Kaldi project about the existing example using the Tedlium dataset. Every step works well until the clean-up stage. I have a length mismatch issue. After examing all the scripts, I found the issue is in the…
chuhui chen
  • 57
  • 5
0
votes
1 answer

copy .wav audio file settings to new .wav file

currently I am working with a speech to text translation model that takes a .wav file and turns the audible speech within the audio into a text transcript. The model worked before on .wav audio recordings that were recorded directly. However now I…
Jonas
  • 67
  • 9
0
votes
2 answers

Install Kaldi permanently on colab

I am new to speech recognition and I wish to build an end-to-end asr system using kaldi-asr. My laptop does not have Linux and neither does it have enough hardware to train models. So, I planned to install Kaldi on colab( to leverage Free GPU) by…
Jor_El
  • 187
  • 8
0
votes
1 answer

Installing Kaldi on MacOS Catalina -- error with zlib

I am trying to work with the DiscVoice Library which requires the Kaldi Library. In order to install Kaldi, I needed to run extras/check_dependencies.sh to check the dependencies of the program, and I am currently…
Sruthi Kurada
  • 21
  • 4
0
votes
0 answers

How to save multiple audio files from one audio file based on speaker?

I have an audio file in which there are 2 speakers, the first one recites a sentence & then the other one translates it. I want to save an audio file for each sentence recited by speaker A & B. Example: file -> book_translation.mp3 ( 1minute audio…
Lakhani Aliraza
  • 435
  • 6
  • 8
0
votes
3 answers

intel mkl fatal error when running kaldi gst live demo

the output errori tried running GST_PLUGIN_PATH=../../src/gst-plugin ./run-live.py and get intel mkl fatal error Intel MKL FATAL ERROR: cannot load libmkl_vml_avx2.so or libmkl_vml_def.so. i tried export…
saudmx3
  • 13
  • 2
0
votes
0 answers

Need phonectics in Kaldi decoding

I am using aspire example in KALDI. And it's working fine. Just there is one help that I need and it is that I need phonetics of an audio file while decoding. So basically when I decode a .wav file I need phonetics of audio instead of the…
Akshay Jain
  • 7
  • 2
0
votes
0 answers

Save the output of a trained pytorch neural network in .ark format

I am training my acoustic model which is multilayer DNN written in PyTorch. I want to save the output of the network as (.ark) file to use it in the decoding stage for automatic speech recognition. I searched a lot but can't find a function to do…
Mohamed Nabih
  • 93
  • 1
  • 8
0
votes
0 answers

Kaldi: Transfer learning

I'm trying to do transfer learning on Kaldi-ASR with a model that has been pretrained on Common Voice, with a custom limited vocabulary dataset. However, I have found the documentation to be quite limited wrt the config file to be provided to…
Kalpit
  • 891
  • 1
  • 8
  • 24
0
votes
0 answers

Kaldi_io error reading .scp/.ark files with python

I need to read some scp file with the following code: for key,vec in kaldi_io.read_vec_flt_scp('raw_mfcc_dev_clean.1.scp'): print("mfcc of id ",key,"is equal to ",vec) here's a line in the scp file: 1272-128104-0000…
user12830038
0
votes
1 answer

training model in english, but evaluating in another language for speech diarization task?

For speech diarization task, can i just train my model on english-based dataset (utterances of single words), but evaluate in my language? Or this does not make sense, and the model will show poor results? I need to implement a rather simple model,…
ERJAN
  • 23,696
  • 23
  • 72
  • 146
0
votes
1 answer

Connection to websocket via ip address refused by firefox

I want to establish a websocket connection to my own server. The code I tried was the following one: var connection = new WebSocket('ws://137.250.171.212:8888/ws/speech'); But I get an error Firefox kann keine Verbindung zu dem Server unter…
AndiLeni
  • 481
  • 1
  • 6
  • 16
1 2 3 4 5
6
7 8