Thanks for all the information Nickolay. You've been very helpful. Vosk does seem to be a very good SRE, but the dictionary modifications appears to be too complex for my purpose. So I'm going to need to go back to using pocketsphinx just for the ease-of-use of its dictionary. In any case, thanks for your patience and time. You've been very helpful.
It can be difficult to get a simple answer to a question. I've got vosk installed and running on a computer. Vosk only. Not Kaldi. Although Vosk apparently comes with some form of a Kaldirecognitzer. My question is where do I find the vocabulary dictionary, and how do I modfied it? And is it easy to do? With pocketsphinx it is extremely easy to modify the dictionary or even created a new dictionary from scratch. And that was the main attraction to pocketsphinx. I need to be able to create a dictionary...
"There is C documentation here. https://cmusphinx.github.io/doc/pocketsphinx/files.html. There is no Python documentation, just the source code." This will be very helpful. Thank you! Thanks for the information on alsa config, and json too. Sorry to hear that there's no documentation for vosk. How's a person supposed to learn it? Just keep coming here and asking questions? I just want to iron out a few things so I can use an SRE for my Linquistic AI project. I like Pocket Sphinx because I understand...
Ok I finally got VOSK running, but as I suspected I have simlar questions with vosk: Here's my Python code: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37#!/usr/bin/python3 from vosk import Model, KaldiRecognizer import time import os os.system('clear') time.sleep(1) if not os.path.exists("model-en"): print ("Please download the model from https://github.com/alphacep/kaldi-android-demo/releases and unpack as 'model-en' in the current folder.")...
I did install libgfortran as a dependency Kaldi. I'll install libgfortran3 and see if that helps. Although I'm probably going to have as many questions about vosk as I have for pocketsphinx. The first thing I'll want to do is empty its dictionary and start my own dictionary from scratch. I already know how to do that for pocketsphinx. Currently I have pocketsphinx running. And it's decoding my speech close to 100% accuracy. In fact, most of the time it is 100% accurate. As I say, it must like my...
I'm not getting anywhere with vosk. I've installed vosk and Kaldi. I'm trying to run the following python text code for vosk 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29#!/usr/bin/python3 from vosk import Model, KaldiRecognizer import sys import os import wave if not os.path.exists("model-en"): print ("Please download the model from https://github.com/alphacep/kaldi-android-demo/releases and unpack as 'model-en' in the current folder.") exit (1) wf = wave.open(sys.argv[1],...
Ok, that seemed to work. james@james-desktop:~/vosk-kaldi$ pip3 install https://github.com/alphacep/vosk-api/releases/download/0.3.3/vosk-0.3.3-cp36-cp36m-linux_aarch64.whl Defaulting to user installation because normal site-packages is not writeable Collecting vosk==0.3.3 Downloading https://github.com/alphacep/vosk-api/releases/download/0.3.3/vosk-0.3.3-cp36-cp36m-linux_aarch64.whl (2.5 MB) |████████████████████████████████| 2.5 MB 4.9 kB/s Installing collected packages: vosk Successfully installed...
Ok, I have a problem right off the bat: I'm running Ubuntu 18.04 on a Jetson Nano arm64. From the GitHub page I tried the foillowing: james@james-desktop:~/vosk-kaldi$ pip3 --version pip 20.0.2 from /home/james/.local/lib/python3.6/site-packages/pip (python 3.6) (I have the correct verion of pip) james@james-desktop:~/vosk-kaldi$ pip3 install vosk Defaulting to user installation because normal site-packages is not writeable ERROR: Could not find a version that satisfies the requirement vosk (from...