Voice Based Gender Detection on Edge for Leading Semiconductor Company in US
Overview
The client is a leading US-based semiconductor manufacturer offering microcontrollers and processors to sensors, analog ICs, and connectivity solutions. They focused on introducing a new line of chipsets designed for Machine Learning at the edge, and they sought to bring innovative applications to the market. ACL Digital designed a voice-based gender detection application. This application utilized a supervised machine learning algorithm, specifically the Depth-wise Separable Convolutional Neural Network. The model was developed using the TensorFlow framework, and the training process involved leveraging extracted features to enhance the gender detection functionality.
Download Case Study
Challenges
Lack of applications for newly developed Machine Learning chipsets
Solutions
- Designed the voice-based gender detection application using supervised machine learning algorithm– Depth-wise separable convolution neural network
- DS-CNN is based on MobileNet architecture that is best suited for memory constrained devices
- Extracted audio features using Mel Frequency Cepstral Coefficients technique
- Developed model using TensorFlow framework and trained model with extracted features from 9000 audio files
- Developed application on NXP i.MXRT600, performed data pre-processing and audio feature extraction (MFCC) of the real-time audio samples on Hi-Fi 4 DSP and the inference of the Deep Learning model was deployed on ARM Cortex-M33
- Performance evaluation using real human voice input through microphone
Outcomes
- Accelerated client’s product launch timeline by 20% with years of expertise in machine learning domain
