Mel Spectrogram Research Articles

Self-supervised learning–based underwater acoustical signal classification via mask modeling

The classification of underwater acoustic signals has garnered a great deal of attention in recent years due to its potential applications in military and civilian contexts. While deep neural networks have emerged as the preferred method for this task, the representation of the signals plays a crucial role in determining the performance of the classification. However, the representation of underwater acoustic signals remains an under-explored area. In addition, the annotation of large-scale datasets for the training of deep networks is a challenging and expensive task. To tackle these challenges, we propose a novel self-supervised representation learning method for underwater acoustic signal classification. Our approach consists of two stages: a pretext learning stage using unlabeled data and a downstream fine-tuning stage using a small amount of labeled data. The pretext learning stage involves randomly masking the log Mel spectrogram and reconstructing the masked part using the Swin Transformer architecture. This allows us to learn a general representation of the acoustic signal. Our method achieves a classification accuracy of 80.22% on the DeepShip dataset, outperforming or matching previous competitive methods. Furthermore, our classification method demonstrates good performance in low signal-to-noise ratio or few-shot settings.

Analysis of Application and Creation Skills of Story-Based MV Micro Video and Big Multimedia Data in Music Communication

MV is an art type that uses TV pictures to supplement information and content that cannot be covered by music. With the rapid development of micro video technology and network technology, micro video has been rapidly popularized. Combining MV and micro video with mobile devices and platforms, it breaks the limitations of traditional movies, TV, and animation and becomes a new highlight of cultural communication in the new era. This paper starts with the selection of MV songs, story script, content design, MV shooting and editing skills, and introduces the specific creation methods of story MV in combination with local culture and proposes a music genre classification algorithm model DCNN-SSA based on spectral space domain feature concern. DCNN-SSA model effectively marks genre features of different music Mel spectrograms in the spatial domain and changes the network structure, thus improving the feature extraction effect while ensuring the validity of the model, thus improving the accuracy of music genre classification.

Cognitive Load Assessment of Air Traffic Controller Based on SCNN-TransE Network Using Speech Data

Due to increased air traffic flow, air traffic controllers (ATCs) operate in a state of high load or even overload for long periods of time, which can seriously affect the reliability and efficiency of controllers’ commands. Thus, the early identification of ATCs who are overworked is crucial to the maintenance of flight safety while increasing overall flight efficiency. This study uses a comprehensive comparison of existing cognitive load assessment methods combined with the characteristics of the ATC as a basis from which a method for the utilization of speech parameters to assess cognitive load is proposed. This method is ultimately selected due to the minimal interference of the collection equipment and the abundance of speech signals. The speech signal is pre-processed to generate a Mel spectrogram, which contains temporal information in addition to energy, tone, and other spatial information. Therefore, a speech cognitive load evaluation model based on a stacked convolutional neural network (CNN) and the Transformer encoder (SCNN-TransE) is proposed. The use of a CNN and the Transformer encoder allows us to extract spatial features and temporal features, respectively, from contextual information from speech data and facilitates the fusion of spatial features and temporal features into spatio-temporal features, which improves our method’s ability to capture the depth features of speech. We conduct experiments on air traffic control communication data, which show that the detection accuracy and F1 score of SCNN-TransE are better than the results from the support-vector machine (SVM), k-nearest neighbors (KNN), random forest (RF), adaptive boosting (AdaBoost), and stacked CNN parallel long short-term memory with attention (SCNN-LSTM-Attention) models, reaching values of 97.48% and 97.07%, respectively. Thus, our proposed model can realize the effective evaluation of cognitive load levels.

Assamese Dialect Identification System using Convolution Neural Networks

Labeling speech in an audio file with appropriate dialect labels is the aim of a dialect identification system. This paper presents a method of using convolution neural networks (CNN) to identify four Assamese dialects: Goalporia dialect, Kamrupi dialect, Eastern Assamese dialect, and Central Assamese dialect. This study employed the speech patterns of four major Assamese regional dialects: the Central Dialects spoken in and around the district of Nagaon; the Eastern Assamese dialect spoken in the districts of Sibsagar and its neighboring areas; the Kamrupi dialect spoken in the districts of Kamrup, Nalbari, Barpeta, Kokarajhar, and some areas of Bongaigaon; and the Goaplari dialect spoken in the Goaplara, Dhuburi, and a portion of Bongaigaon district. Over the course of two hours, audio samples from each of the four dialects were used to train the classifier. Mel spectrogram pictures, which are produced from two to four second divisions of raw audio input with varying audio quality, are used by the CNN. The system's performance is also analyzed in relation to the lengths of the train and test audio samples. The proposed CNN model achieves an accuracy of 90.82 percent, which may be the best when compared to machine learning models.

FPGA Implementation of Keyword Spotting System Using Depthwise Separable Binarized and Ternarized Neural Networks.

Keyword spotting (KWS) systems are used for human-machine communications in various applications. In many cases, KWS involves a combination of wake-up-word (WUW) recognition for device activation and voice command classification tasks. These tasks present a challenge for embedded systems due to the complexity of deep learning algorithms and the need for optimized networks for each application. In this paper, we propose a depthwise separable binarized/ternarized neural network (DS-BTNN) hardware accelerator capable of performing both WUW recognition and command classification on a single device. The design achieves significant area efficiency by redundantly utilizing bitwise operators in the computation of the binarized neural network (BNN) and ternary neural network (TNN). In a complementary metal-oxide semiconductor (CMOS) 40 nm process environment, the DS-BTNN accelerator demonstrated significant efficiency. Compared with a design approach where BNN and TNN were independently developed and subsequently integrated as two separate modules into the system, our method achieved a 49.3% area reduction while yielding an area of 0.558 mm2. The designed KWS system, which was implemented on a Xilinx UltraScale+ ZCU104 field-programmable gate array (FPGA) board, receives real-time data from the microphone, preprocesses them into a mel spectrogram, and uses this as input to the classifier. Depending on the order, the network operates as a BNN or a TNN for WUW recognition and command classification, respectively. Operating at 170 MHz, our system achieved 97.1% accuracy in BNN-based WUW recognition and 90.5% in TNN-based command classification.

Feeding intensity assessment of aquaculture fish using Mel Spectrogram and deep learning algorithms

Accurately and objectively analyzing fish feeding intensity is essential to guiding feeding and production techniques. Fish feeding intensity in recirculating aquaculture systems (RAS) can be used to indicate a fish's appetite. However, the low efficiency and lack of objectivity of manual observation are problems with current fish feeding intensity assessment processes. Applying acoustic techniques to aquaculture issues is an insufficiently explored area that requires new investigations, particularly into methods that explore temporal information in acoustic spectrograms. With Oplegnathus punctatus as the experimental species, we proposed a fish feeding intensity detection method based on the Mel Spectrogram and MobileNetV3-SBSC lightweight networks. First, the Oplegnathus punctatus feeding sound dataset, which has a total of 3 types—"strong," "medium," and "none," was built. Next, Mel Spectrogram feature maps were extracted using steps including preprocessing, Fast Fourier Transform (FFT), Mel filter bank (MFB), etc. Finally, the MobileNetV3-SBSC lightweight network was used to detect and recognize the obtained feature maps. Experimental results indicated that the proposed MobileNetV3-SBSC model, as compared to the MobileNetV3-S model, improved test accuracy by 4.6% and decreased test loss by 67.4% with only a 0.84% increase in the number of parameters and a 3.08% increase in the model size. Additionally, we have verified that the accuracy of the test set was 59.6%, 53.3%, 83.3%, 85.3%, and 85.9% for groups of 5, 15, 40, 70, and 100 fish, respectively. This study demonstrated that the proposed method is not applicable to a small number of fish, which means that when the numbers of fish are small, changes in the feeding of individual fish would have a significant effect on the whole feeding feature map, leading to negligible changes in feeding features. However, in view of the commonly high aquaculture density, the proposed method can be used to automatically and objectively examine fish feeding, which could provide a theoretical basis and methodological support for further feeding decisions.

Hybrid data augmentation and deep attention-based dilated convolutional-recurrent neural networks for speech emotion recognition

Recently, speech emotion recognition (SER) has become an active research area in speech processing, particularly with the advent of deep learning (DL). Numerous DL-based methods have been proposed for SER. However, most of the existing DL-based models are complex and require a large amounts of data to achieve a good performance. In this study, a new framework of deep attention-based dilated convolutional-recurrent neural networks coupled with a hybrid data augmentation method was proposed for addressing SER tasks. The hybrid data augmentation method constitutes an upsampling technique for generating more speech data samples based on the traditional and generative adversarial network approaches. By leveraging both convolutional and recurrent neural networks in a dilated form along with an attention mechanism, the proposed DL framework can extract high-level representations from three-dimensional log Mel spectrogram features. Dilated convolutional neural networks acquire larger receptive fields, whereas dilated recurrent neural networks overcome complex dependencies as well as the vanishing and exploding gradient issues. Furthermore, the loss functions are reconfigured by combining the SoftMax loss and the center-based losses to classify various emotional states. The proposed framework was implemented using the Python programming language and the TensorFlow deep learning library. To validate the proposed framework, the EmoDB and ERC benchmark datasets, which are imbalanced and/or small datasets, were employed. The experimental results indicate that the proposed framework outperforms other related state-of-the-art methods, yielding the highest unweighted recall rates of 88.03 ± 1.39 (%) and 66.56 ± 0.67 (%) for the EmoDB and ERC datasets, respectively.

Voice pathology identification system using a deep learning approach based on unique feature selection sets

AbstractVoice pathology diagnosis requires extracting significant features from voice signals, and classical machine learning models can overfit to the training data, which can cause difficult issues and pose challenges. The study aimed to develop a reliable and efficient system for identifying voice pathologies utilizing the long short‐term memory (LSTM) method. The study combined unique feature sets such as the mel frequency cepstral coefficients (MFCCs), zero crossing rate (ZCR), and mel spectrograms, which have not been used together in previous works. Voice pathology identification improved the accuracy rate using the LSTM approach on the Saarbruecken voice database (SVD) samples. The best results achieved by the proposed system showed an accuracy rate of 99.3% for /u/ vowel samples in neutral pitch, 99.2% for /a/ vowel samples in high pitch, 99% for /i/ vowel samples in neutral pitch, and 99.2% for sentence samples. The experimental results were evaluated utilizing accuracy, precision, specificity, sensitivity, and F1 measures. Additionally, the study compared the performance of LSTM with that of artificial neural networks (ANNs) and found that LSTM achieved better outcomes.

Multichannel CNN-BLSTM Architecture for Speech Emotion Recognition System by Fusion of Magnitude and Phase Spectral Features Using DCCA for Consumer Applications

Conventional Speech Emotion Recognition (SER) approaches put more emphasis on extracting magnitude spectrum-based features, such as Mel Frequency Cepstral Coefficients (MFCCs), and Mel spectrogram. However, phase information is ignored due to signal processing difficulties such as the phase wrapping issue. This work develops a multichannel Convolution Neural Network-Bidirectional Long Short Term Memory (CNN-BLSTM) architectures with an attention mechanism for speaker-independent SER by considering phase and magnitude spectrum-based features. The phase-based features are extracted using the Modified Group Delay Function (MODGD). The obtained phase features are combined with MFCC features. The CNN-BLSTM network extract learned representation from magnitude and phase features. The learned representation from MFCCs and MODGD are combined and given as an input to the Support Vector Machine (SVM) for classification. The Deep Canonical Correlation Analysis (DCCA) is used to maximize the correlation between magnitude and phase information to improve the conventional SER system’s performance. Here the IEMOCAP database is used for performance analysis. The experimental results show improvement over MFCC features and existing approaches for unimodal SER. In this work, we also developed real-time web server application for the proposed architecture.

E-DGAN: An Encoder-Decoder Generative Adversarial Network Based Method for Pathological to Normal Voice Conversion.

In recent years, more and more people suffer from voice-related diseases. Given the limitations of current pathological speech conversion methods, that is, a method can only convert a single kind of pathological voice. In this study, we propose a novel Encoder-Decoder Generative Adversarial Network (E-DGAN) to generate personalized speech for pathological to normal voice conversion, which is suitable for multiple kinds of pathological voices. Our proposed method can also solve the problem of improving the intelligibility and personalizing custom speech of pathological voices. Feature extraction is performed using a mel filter bank. The conversion network is an encoder-decoder structure, which is used to convert the mel spectrogram of pathological voices to the mel spectrogram of normal voices. After being converted by the residual conversion network, the personalized normal speech is synthesized by the neural vocoder. In addition, we propose a subjective evaluation metric named "content similarity" to evaluate the consistency between the converted pathological voice content and the reference content. The Saarbrücken Voice Database (SVD) is used to verify the proposed method. The intelligibility and content similarity of pathological voices are increased by 18.67% and 2.60%, respectively. Besides, an intuitive analysis based on a spectrogram was done and a significant improvement was achieved. The results show that our proposed method can improve the intelligibility of pathological voices and personalize the conversion of pathological voices into the normal voices of 20 different speakers. Our proposed method is compared with five other pathological voice conversion methods, and our proposed method has the best evaluation results.

CNN-Based Identification of Parkinson's Disease from Continuous Speech in Noisy Environments.

Parkinson's disease is a progressive neurodegenerative disorder caused by dopaminergic neuron degeneration. Parkinsonian speech impairment is one of the earliest presentations of the disease and, along with tremor, is suitable for pre-diagnosis. It is defined by hypokinetic dysarthria and accounts for respiratory, phonatory, articulatory, and prosodic manifestations. The topic of this article targets artificial-intelligence-based identification of Parkinson's disease from continuous speech recorded in a noisy environment. The novelty of this work is twofold. First, the proposed assessment workflow performed speech analysis on samples of continuous speech. Second, we analyzed and quantified Wiener filter applicability for speech denoising in the context of Parkinsonian speech identification. We argue that the Parkinsonian features of loudness, intonation, phonation, prosody, and articulation are contained in the speech, speech energy, and Mel spectrograms. Thus, the proposed workflow follows a feature-based speech assessment to determine the feature variation ranges, followed by speech classification using convolutional neural networks. We report the best classification accuracies of 96% on speech energy, 93% on speech, and 92% on Mel spectrograms. We conclude that the Wiener filter improves both feature-based analysis and convolutional-neural-network-based classification performances.

Speech Emotion Recognition

Emotions are the best way for people to communicate their thoughts and actions to others. The most important technology in the world today is the ability to recognize emotions from a single speaker's voice. The ability to recognize emotions is very useful in gaining various insightful insights into a person's thoughts. The process of extracting emotions from human speech is called Speech Emotion Recognition (SER). We used the RAVDESS (Ryerson AudioVisual Database of Emotional Speech and Song) dataset to extract emotions from Speech. Emotions are extracted from speech based on speech parameters such as Mel-FrequencyCepstral -Coefficients (MFCC) and Mel Spectrogram. After training with a Multilayer Perceptron classifier (MLP), the obtained data had an accuracy of 68.33% and accuracy of 80.64% after training with Convolutional Neural Networks Long Short Term Memory (CNN LSTM).

Comparative Study of Mfcc and Mel Spectrogram for Raga Classification Using CNN

Objectives: To perform a comparative study of the results of feature extraction done using two different methods, Mfcc and Mel spectrogram, and determine which method is more effective for implementing the CNN algorithm. Methods: This study uses the CNN model to classify ragas according to Indian classical music. Feature extraction, which is a major operation in the Music Information Retrieval (MIR) process, is done using Mfcc and Mel spectrogram methods. The major ragas chosen as subjects for feature extraction are Yaman, Bhairav, Bhairavi, Multani, and Dhanashree. Findings: After comparison and examination of results achieved from both techniques, we could conclude that the CNN model using the Mel spectrogram method outperforms the CNN model using Mfcc. Novelty: The majority of the research, we discovered was on Carnatic music. In contrast to earlier research, this research takes a novel approach by conducting experiments on a variety of Hindustani classical ragas which are different from other studies. Researchers interested in music as well as application users would benefit from this study. Our proposed feature extraction approach will be useful for initializing the CNN algorithm, which will help aspiring musicians recognize ragas and classify songs based on these ragas. Keywords: Raga identification; Music Information Retrieval; Feature extraction; Mfcc; Melspectrograms; CNN

Music sentiment classification based on an optimized CNN-RF-QPSO model

PurposeMusic sentiment analysis helps to promote the diversification of music information retrieval methods. Traditional music emotion classification tasks suffer from high manual workload and low classification accuracy caused by difficulty in feature extraction and inaccurate manual determination of hyperparameter. In this paper, the authors propose an optimized convolution neural network-random forest (CNN-RF) model for music sentiment classification which is capable of optimizing the manually selected hyperparameters to improve the accuracy of music sentiment classification and reduce labor costs and human classification errors.Design/methodology/approachA CNN-RF music sentiment classification model is designed based on quantum particle swarm optimization (QPSO). First, the audio data are transformed into a Mel spectrogram, and feature extraction is conducted by a CNN. Second, the music features extracted are processed by RF algorithm to complete a preliminary emotion classification. Finally, to select the suitable hyperparameters for a CNN, the QPSO algorithm is adopted to extract the best hyperparameters and obtain the final classification results.FindingsThe model has gone through experimental validations and achieved a classification accuracy of 97 per cent for different sentiment categories with shortened training time. The proposed method with QPSO achieved 1.2 and 1.6 per cent higher accuracy than that with particle swarm optimization and genetic algorithm, respectively. The proposed model had great potential for music sentiment classification.Originality/valueThe dual contribution of this work comprises the proposed model which integrated two deep learning models and the introduction of a QPSO into model optimization. With these two innovations, the efficiency and accuracy of music emotion recognition and classification have been significantly improved.

Estimation of subjective quality of life in schizophrenic patients using speech features.

Patients with schizophrenia experience the most prolonged hospital stay in Japan. Also, the high re-hospitalization rate affects their quality of life (QoL). Despite being an effective predictor of treatment, QoL has not been widely utilized due to time constraints and lack of interest. As such, this study aimed to estimate the schizophrenic patients' subjective quality of life using speech features. Specifically, this study uses speech from patients with schizophrenia to estimate the subscale scores, which measure the subjective QoL of the patients. The objectives were to (1) estimate the subscale scores from different patients or cross-sectional measurements, and 2) estimate the subscale scores from the same patient in different periods or longitudinal measurements. A conversational agent was built to record the responses of 18 schizophrenic patients on the Japanese Schizophrenia Quality of Life Scale (JSQLS) with three subscales: "Psychosocial," "Motivation and Energy," and "Symptoms and Side-effects." These three subscales were used as objective variables. On the other hand, the speech features during measurement (Chromagram, Mel spectrogram, Mel-Frequency Cepstrum Coefficient) were used as explanatory variables. For the first objective, a trained model estimated the subscale scores for the 18 subjects using the Nested Cross-validation (CV) method. For the second objective, six of the 18 subjects were measured twice. Then, another trained model estimated the subscale scores for the second time using the 18 subjects' data as training data. Ten different machine learning algorithms were used in this study, and the errors of the learned models were compared. The results showed that the mean RMSE of the cross-sectional measurement was 13.433, with k-Nearest Neighbors as the best model. Meanwhile, the mean RMSE of the longitudinal measurement was 13.301, using Random Forest as the best. RMSE of less than 10 suggests that the estimated subscale scores using speech features were close to the actual JSQLS subscale scores. Ten out of 18 subjects were estimated with an RMSE of less than 10 for cross-sectional measurement. Meanwhile, five out of six had the same observation for longitudinal measurement. Future studies using a larger number of subjects and the development of more personalized models based on longitudinal measurements are needed to apply the results to telemedicine for continuous monitoring of QoL.

EEG driving fatigue detection based on log-Mel spectrogram and convolutional recurrent neural networks.

Driver fatigue detection is one of the essential tools to reduce accidents and improve traffic safety. Its main challenge lies in the problem of how to identify the driver's fatigue state accurately. Existing detection methods include yawning and blinking based on facial expressions and physiological signals. Still, lighting and the environment affect the detection results based on facial expressions. In contrast, the electroencephalographic (EEG) signal is a physiological signal that directly responds to the human mental state, thus reducing the impact on the detection results. This paper proposes a log-Mel spectrogram and Convolution Recurrent Neural Network (CRNN) model based on EEG to implement driver fatigue detection. This structure allows the advantages of the different networks to be exploited to overcome the disadvantages of using them individually. The process is as follows: first, the original EEG signal is subjected to a one-dimensional convolution method to achieve a Short Time Fourier Transform (STFT) and passed through a Mel filter bank to obtain a logarithmic Mel spectrogram, and then the resulting logarithmic Mel spectrogram is fed into a fatigue detection model to complete the fatigue detection task for the EEG signals. The fatigue detection model consists of a 6-layer convolutional neural network (CNN), bi-directional recurrent neural networks (Bi-RNNs), and a classifier. In the modeling phase, spectrogram features are transported to the 6-layer CNN to automatically learn high-level features, thereby extracting temporal features in the bi-directional RNN to obtain spectrogram-temporal information. Finally, the alert or fatigue state is obtained by a classifier consisting of a fully connected layer, a ReLU activation function, and a softmax function. Experiments were conducted on publicly available datasets in this study. The results show that the method can accurately distinguish between alert and fatigue states with high stability. In addition, the performance of four existing methods was compared with the results of the proposed method, all of which showed that the proposed method could achieve the best results so far.

Multi-modal approach for COVID-19 detection using coughs and self-reported symptoms

COVID-19 (Coronavirus Disease of 2019) is one of the most challenging healthcare crises of the twenty-first century. The pandemic causes many negative impacts on all aspects of life and livelihoods. Although recent developments of relevant vaccines, such as Pfizer/BioNTech mRNA, AstraZeneca, or Moderna, the emergence of new virus mutations and their fast infection rate yet pose significant threats to public health. In this context, early detection of the disease is an important factor to reduce its effect and quickly control the spread of pandemic. Nevertheless, many countries still rely on methods that are either expensive and time-consuming (i.e., Reverse-transcription polymerase chain reaction) or uncomfortable and difficult for self-testing (i.e., Rapid Antigen Test Nasal). Recently, deep learning methods have been proposed as a potential solution for COVID-19 analysis. However, previous works usually focus on a single symptom, which can omit critical information for disease diagnosis. Therefore, in this study, we propose a multi-modal method to detect COVID-19 using cough sounds and self-reported symptoms. The proposed method consists of five neural networks to deal with different input features, including CNN-biLSTM for MFCC features, EfficientNetV2 for Mel spectrogram images, MLP for self-reported symptoms, C-YAMNet for cough detection, and RNNoise for noise-canceling. Experimental results demonstrated that our method outperformed the other state-of-the-art methods with a high AUC, accuracy, and F1-score of 98.6%, 96.9%, and 96.9% on the testing set.

Mel frequency spectral domain defenses against adversarial attacks on speech recognition systems

Automatic speech recognition (ASR) systems are vulnerable to adversarial attacks due to their reliance on machine learning models. Many of the defenses explored for defending ASR systems simply adapt defense approaches developed for the image domain. This paper explores speech-specific defenses in the feature domain and introduces a defense method called mel domain noise flooding (MDNF). MDNF injects additive noise to the mel spectrogram speech representation prior to re-synthesizing the audio signal input to ASR. The defense is evaluated against strong white-box threat models and shows competitive robustness.

Multi-Input Speech Emotion Recognition Model Using Mel Spectrogram and GeMAPS.

The existing research on emotion recognition commonly uses mel spectrogram (MelSpec) and Geneva minimalistic acoustic parameter set (GeMAPS) as acoustic parameters to learn the audio features. MelSpec can represent the time-series variations of each frequency but cannot manage multiple types of audio features. On the other hand, GeMAPS can handle multiple audio features but fails to provide information on their time-series variations. Thus, this study proposes a speech emotion recognition model based on a multi-input deep neural network that simultaneously learns these two audio features. The proposed model comprises three parts, specifically, for learning MelSpec in image format, learning GeMAPS in vector format, and integrating them to predict the emotion. Additionally, a focal loss function is introduced to address the imbalanced data problem among the emotion classes. The results of the recognition experiments demonstrate weighted and unweighted accuracies of 0.6657 and 0.6149, respectively, which are higher than or comparable to those of the existing state-of-the-art methods. Overall, the proposed model significantly improves the recognition accuracy of the emotion "happiness", which has been difficult to identify in previous studies owing to limited data. Therefore, the proposed model can effectively recognize emotions from speech and can be applied for practical purposes with future development.

Influence of several audio parameters in urban sound event classification

This paper presents a system for recognition and classification of sound events established through the development of processes for detection and parameterization of sound signals that are used for further classification by applying machine learning algorithms. The focus is on the feature extraction process and the accuracy achieved when using different audio parameters. Five audio parameters were analyzed and tested on three machine learning algorithms order to investigate their influence in recognizing the class of the disturbing sound events in the urban areas, thus yielding to a collection of 48 different combinations. The applied audio parameters: MFCC, Mel Spectrogram, Chromagram, Spectral Contrast and Tonal Centroid were chosen as they are widely used for urban noise classification, but their combination forming diverse feature vectors have not been still analyzed. The accuracy results were different for each combination of the audio parameters, thus leading to choosing best set of parameters that reach the highest recognition accuracy. The used algorithms: Random Forest (RF), Naïve Bayes (NB) and Support Vector Machines (SVM) classifier are chosen as they have shown very efficient classification in the field of urban noise recognition.