---
license: mit
language:
- sr
metrics:
- wer
base_model:
- facebook/wav2vec2-large-xlsr-53
pipeline_tag: audio-classification
tags:
- voice-command
- smart-home
- voice-controller
- serbian-wav2vec
- audio-classification
---
# Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

Smart home controller simulator, receiving voice commands from a microphone.
Trained to detect the words: "vrata", "svetlo", "zvuk", "otvori", "zatvori", "uključi" and "isključi" to control the state of door, lights and audio in a smart home system.

## Model Details


### Model Description


- **Developed by:** [Mihailo Radović](https://www.linkedin.com/in/mihailo-radović-484070278/)
- **Model type:** Audio Classification (Smart home controller)
- **Language(s) (NLP):** Serbian
- **License:** MIT
- **Finetuned from model:** facebook/wav2vec2-large-xlsr-53

### Model Sources

<!-- Provide the basic links for the model. -->

- **Repository:** [GitHub Repo](https://github.com/mradovic38/voice-command-recognition)
- **Demo:** [Streamlit App](https://smart-home-serbian-voice-controller.streamlit.app)

## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

### Direct Use

Detecting a word said in a short audio clip: "vrata", "svetlo", "zvuk", "otvori", "zatvori", "uključi" and "isključi" to control the state of door, lights and audio in a smart home system.


### Out-of-Scope Use

Works the best for the words that are in the dataset. For the words that are out of the vocabulary, the DTW check is implemented (see code in [GitHub Repo](https://github.com/mradovic38/voice-command-recognition)).


## How to Get Started with the Model

Check out the explaination in my [GitHub repository](https://github.com/mradovic38/voice-command-recognition) README file.