Popular GitHub Repositories for Cepstral

Front-end speech processing aims at extracting proper features from short- term segments of a speech utterance, known as frames. It is a pre-requisite step toward any pattern recognition problem employing speech or audio (e.g., music). Here, we are interesting in voice disorder classification. That is, to develop two-class classifiers, which can discriminate between utterances of a subject suffering from say vocal fold paralysis and utterances of a healthy subject.The mathematical modeling of the speech production system in humans suggests that an all-pole system function is justified [1-3]. As a consequence, linear prediction coefficients (LPCs) constitute a first choice for modeling the magnitute of the short-term spectrum of speech. LPC-derived cepstral coefficients are guaranteed to discriminate between the system (e.g., vocal tract)

Traditional ASR (Signal & Cepstral Analysis, DTW, HMM) & DNNs (Custom Models + DeepSpeech) on Indian Accent Speech

:sound: :boy: :girl: :woman: :man: Speaker identification using voice MFCCs and GMM

一个 Python 库，用于计算两个输入之间的梅尔倒谱距离（梅尔倒谱失真，MCD）。此实现基于 Robert F. Kubichek 在“用于客观语音质量评估的梅尔倒谱距离度量”中提出的方法。

一个用于计算两个输入之间梅尔倒谱距离 (Mel-Cepstral Distortion, MCD) 的 Python 库。此实现基于 Robert F. Kubichek 在“用于客观语音质量评估的梅尔倒谱距离度量”中提出的方法。

Deep Learning model for lexical stress detection in spoken English

In this work we propose two postprocessing approaches applying convolutional neural networks (CNNs) either in the time domain or the cepstral domain to enhance the coded speech without any modification of the codecs. The time domain approach follows an end-to-end fashion, while the cepstral domain approach uses analysis-synthesis with cepstral domain features. The proposed postprocessors in both domains are evaluated for various narrowband and wideband speech codecs in a wide range of conditions. The proposed postprocessor improves speech quality (PESQ) by up to 0.25 MOS-LQO points for G.711, 0.30 points for G.726, 0.82 points for G.722, and 0.26 points for adaptive multirate wideband codec (AMR-WB). In a subjective CCR listening test, the proposed postprocessor on G.711-coded speech exceeds the speech quality of an ITU-T-standardized pos

Classifying AI Synthesised Voice and Human Voice using Machine Learning by Spectral and Cepstral Analysis. Also classified different TTS(Text-to-Speech) engines for different AI synthesized Voice. Published Paper for the whole art of work. Link Given below.

E3WS: Earthquake Early Warning starting from 3 seconds of records on a single station with machine learning

This project explores emotion recognition in audio data, focusing on feature extraction techniques while also comparing the performance of LSTM and 1D CNN models.

Emotion Recognition in Speech: This project leverages advanced machine learning techniques to classify emotions from speech using the Toronto Emotional Speech Set (TESS). By extracting Mel-Frequency Cepstral Coefficients (MFCC) and utilizing an LSTM-based deep learning model, the project accurately identifies emotions like anger, happiness, and sad