Android voice recognition application with multi speaker feature

Abstract

In this paper, we discuss two things. In the first part of the paper, we discuss the means of processing the speech signal using the Mel Frequency Cepstral Coefficients and the basics of voice characteristics. The problem of voice recognition was already implemented at the hardware level, but in this we propose a solution to the problem on the software end with a higher level of accuracy. Then in the second part, we will discuss the application developed on Android which is able to perform both single speaker and multi-speaker recognition. The multispeaker feature enables the system to identify more than one person speaking by splitting the input into smaller blocks and treating each block as a single speaker identification problem. In the single speaker recognition case, the algorithm is able to correctly identify the individual 90.3% of the time. In the multispeaker recognition case, up to 3 individuals are correctly identified roughly 86% of the time. The average identification time is 220ms across a database of 415 samples.

Orchestrating the Development Lifecycle of Machine Learning-based IoT Applications

“…An improved RNN named Long Short Term Memory (LSTM) [147], including complex gates and memory cells within the hidden units for "better memories", became popular in various applications such as speech recognition [126], video analysis [335], language translation [215], activity recognition [130] etc. Since data streaming is most common in the IoT environment, RNN (LSTM) is deemed as one of the most powerful modelling techniques, and there are various IoT applications such as smart assistant [109,336], smart car navigator system [161], malware threat hunting [134], network traffic forecasting [272], equipment condition forecasting [387], energy demand prediction system [243], load forecasting [181], etc.…”

Orchestrating the Development Lifecycle of Machine Learning-Based IoT Applications: A Taxonomy and Survey

“…etc. Since data streaming the most common form in the IoT environment, RNN(LSTM) is deemed as one of the most powerful modelling techniques, and there are various IoT applications such as smart assistant [82,266], smart car navigator system [119], malware threat hunting [101], network traffic forecasting [208], equipment condition forecasting [304], Energy Demand Prediction system [182], load forecasting [131], etc.…”

Personal and Intelligent Home Assistant to Control Devices Using Raspberry Pi

“…Among these technologies, the best security technology is the biometric recognition technology to be utilized. It can be expected to increase security and convenience by applying Speaker Recognition through this biometric recognition technology [2]. However, Echo cannot capable to cover entire home, this great digital companion works perfect if you are in a room or in the adjacent room.…”