Build an audio training model in Python
Budget: €30 – €250 EUR
I am looking for a freelancer to build an audio training model in Python.
The model will be used to train audio models using my computer's .mp3 audio files. The model will allow me to input N clusters, each of which will include N .mp3 audio files. Once the model is trained and tested using a 20% of the dataset randomly, it will print the accuracy.
It will also allow me to do the inference using other audio files in my computer. Whenever it detects a cluster which is not noise, it will print in terminal the audio offset, the probability (at least 50% to be printed) and the cluster detected name.
Required Skills:
The freelancer should have experience in Python and machine learning. He should also be familiar with audio processing and signal analysis.
Deliverables:
The freelancer will deliver the following deliverables:
* The Python code that implements the audio training model and inference
* A README file that explains how to use the code
Find attached one example of how the dataset is built:
- Please find a >4 minutes .mp3 file including the "noise" in the zip file "Fortnite_gameplay_1_ruido.zip".
- Please find 20 files of 2 seconds each .mp3 including the "kill" sound in the zip file "Fortnite_gameplay_1_kill.zip"
The model will be used to train audio models using my computer's .mp3 audio files. The model will allow me to input N clusters, each of which will include N .mp3 audio files. Once the model is trained and tested using a 20% of the dataset randomly, it will print the accuracy.
It will also allow me to do the inference using other audio files in my computer. Whenever it detects a cluster which is not noise, it will print in terminal the audio offset, the probability (at least 50% to be printed) and the cluster detected name.
Required Skills:
The freelancer should have experience in Python and machine learning. He should also be familiar with audio processing and signal analysis.
Deliverables:
The freelancer will deliver the following deliverables:
* The Python code that implements the audio training model and inference
* A README file that explains how to use the code
Find attached one example of how the dataset is built:
- Please find a >4 minutes .mp3 file including the "noise" in the zip file "Fortnite_gameplay_1_ruido.zip".
- Please find 20 files of 2 seconds each .mp3 including the "kill" sound in the zip file "Fortnite_gameplay_1_kill.zip"