Giter Club home page Giter Club logo

Comments (4)

cgreening avatar cgreening commented on August 13, 2024

If you look in both the Train Model.ipynb and Generate Training Data.ipynb you will see that there is a list of words.

Generate Training Data creates sets of features for all the audio samples for each word in the list and tags them with the index of the word in the list.

So for example backward = 0, bed = 1, bird = 2

The Train Model code marks any features that are not tagged with the same index as the word marvin as negative cases and all the features tagged with the same index as the word marvin as positive cases.

It then trains a binary classifier against this data.

In the firmware we feed the audio data to the trained model and it returns a probability that the word marvin appears in the audio data.

# List of the words in categorical order
words = [
    'backward',
    'bed',
    'bird',
    'cat',
    'dog',
    'down',
    'eight',
    'five',
    'follow',
    'forward',
    'four',
    'go',
    'happy',
    'house',
    'learn',
    'left',
    'marvin',
    'nine',
    'no',
    'off',
    'on',
    'one',
    'right',
    'seven',
    'sheila',
    'six',
    'stop',
    'three',
    'tree',
    'two',
    'up',
    'visual',
    'wow',
    'yes',
    'zero',
    '_background',
]

from diy-alexa.

Anugaradhawa avatar Anugaradhawa commented on August 13, 2024

Thanks, But I didn't receive an Expected answer. Now think I make a new wake word then how do I add this new wake word to code??

from diy-alexa.

cgreening avatar cgreening commented on August 13, 2024

You need to collect a lot of samples of your wake word all of them 1 second long and you need to make sue you use 16KHz sampling rate.

Create a folder and in the speech_data directory and put your samples in there. Copy the name of the folder and put it in the list of words in both notebooks. When the notebook uses "marvin" change it to your wake words.

In the generate training data it repeats the samples for marvin many times to balance the dataset.

In the train model dataset it marks all the marvin words as the positive case.

You will need to learn Python to modify the notebooks. Please read and understand the code. I'm afraid I cannot do your project for you.

from diy-alexa.

Anugaradhawa avatar Anugaradhawa commented on August 13, 2024

Please read and understand the code. I'm afraid I cannot do your project for you.

ok ok I try to make my own wake word thanks for helping for my project

from diy-alexa.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.