Following recent advancements in deep learning and artificial intelligence, spoken language identification applications are playing an increasingly significant role in our day-to-day lives, especially in the domain of multi-lingual speech recognition. In this article, we propose a spoken language identification system that depends on the sequence of feature vectors. The proposed system uses a hybrid Convolutional Recurrent Neural Network (CRNN), which combines a Convolutional Neural Network (CNN) with a Recurrent Neural Network (RNN) network, for spoken language identification on seven languages, including Arabic, chosen from subsets of the Mozilla Common Voice (MCV) corpus. The proposed system exploits the advantages of both CNN and RNN architectures to construct the CRNN architecture. At the feature extraction stage, it compares the Gammatone Cepstral Coefficient (GTCC) feature and Mel Frequency Cepstral Coefficient (MFCC) feature, as well as a combination of both. Finally, the speech signals were represented as frames and used as the input for the CRNN architecture. After conducting experiments, the results of the proposed system indicate higher performance with combined GTCC and MFCC features compared to GTCC or MFCC features used individually. The average accuracy of the proposed system was 92.81% in the best experiment for spoken language identification. Furthermore, the system can learn language-specific patterns in various filter size representations of speech files.

Crowd density estimation is an important topic in computer vision due to its widespread applications in surveillance, urban planning, and intelligence gathering. Resulting from extensive analysis, crowd density estimation reflects many aspects such as similarity of appearance between people, background components, and inter-blocking in intense crowds. In this paper, we are interested to apply machine learning for crowd management in order to monitor populated area and prevent congestion situations. We propose a Single-Convolutional Neural Network with Three Layers (S-CNN3) model to count the number of people in a scene and conclude about the crowd estimation. Then, a comparative study for density counting establishes the performance of the proposed model against the convolutional neural networks with four layers (single-CNN4) and Switched Convolutional neural networks (SCNN). ShanghaiTech dataset, considered as the largest data base for crowd counting, is used in this work. The proposed model proves high effectiveness and efficiency for crowd density estimation with 99.88% of average test accuracy and 0.02 of average validation loss. These results achieve better performance than the existing state-of-the-art models.

The rapid momentum of deep neural networks (DNNs) in recent years has yielded state-of-the-art performance in various machine-learning tasks using speaker identification systems. Speaker identification is based on the speech signals and the features that can be extracted from them. In this article, we proposed a speaker identification system using the developed DNNs models. The system is based on the acoustic and prosodic features of the speech signal, such as pitch frequency (vocal cords vibration rate), energy (loudness of speech), their derivations, and any additional acoustic and prosodic features. Additionally, the article investigates the existing recurrent neural networks (RNNs) models and adapts them to design a speaker identification system using the public YOHO LDC dataset. The average accuracy of the system was 91.93% in the best experiment for speaker identification. Furthermore, this paper helps uncover reasons for analyzing speakers and tokens yielding major errors to increase the system’s robustness regarding feature selection and system tune-up.

The StarGANv2-VC model is a many-to-many non-parallel generative adversarial network (GAN) voice conversion (VC) model that has proven effective in style conversion tasks. This study aimed to investigate the scalability and diversity of the model for English emotional voice conversion (EVC) across different speakers and emotions. We carried out many experiments using an Emotional Speech Database (ESD), comprising a single speaker-multi-emotion experiment, a multi-speakers-multi-emotions experiment (gender-dependent), and a multi-speakers-multi-emotions experiment (gender-independent). We also assessed the effect of training set size and compared the performance of the StarGANv2-VC model with a CycleGAN model. Our study found that the StarGANv2-VC model accurately converted the pitch of the voice across all four emotions (neutral, happy, sad, and angry). However, the model’s efficiency in converting multi-emotions to multi-speakers was not as high as its efficiency in voice conversion for multi-speakers. Further research is needed in this area. We objectively assessed the quality of the converted speech using Mel-frequency cepstral distortion (MCD) and root-mean-square error (RMSE) for spectrum and prosody, respectively. Additionally, we conducted cross-emotion recognition using a convolutional recurrent neural network (CRNN).

Spoken Language Identification (SLID) is an important step in speech-to-speech translation systems and multi-lingual automatic speech recognition. In recent research, deep learning mechanisms have been the prevailing approaches for spoken language identification. This paper aims to study, detect, and analyze spoken languages similar to Arabic in pronouncing certain words and then proposes a deep learning-based architecture, specifically the Bidirectional Long Short Term Memory (BLSTM), for spoken Arabic language identification and discrimination between these similar languages, namely, German, Spanish, French, and Russian, all of which are taken from Mozilla speech corpus languages. Additionally, our work involves a linguistic study of these considered languages. A total of ten thousand speakers are chosen for all five languages, and the BLSTM architecture is designed and implemented using acoustic signal features and applied to five experiments in this paper. The results show a precision of 98.97%, 98.73%, 98.47%, and 99.75% for identifying the spoken Arabic language separately along with German, Spanish, French, and Russian, respectively. Additionally, we achieved an average accuracy of 95.15% for discriminating between all these considered five languages in terms of the pronunciation of words. Our findings confirm that a BLSTM architecture is able to distinguish between observable similar pronunciations of words in considered languages.

Determination of speaker gender from a speech is an important subject, where can be applied to various real-world applications. This study aims to investigate the robustness of classifying gender with different speakers who speak multi-languages, specifically Arabic and English, by using the Bidirectional Long Short-Term Memory (BLSTM) network model-based classifier. For that purpose, this paper presents the mono-language and cross-language gender classification results obtained using the BLSTM model from Arabic and English language datasets were selected from the open-source Mozilla Common Voice (MCV) corpus. The proposed model has mainly based on the speech signal features, including the Gammatone Cepstral Coefficient (GTCC), Spectral Entropy, Pitch, Harmonic Ratio, Mel Frequency Cepstral Coefficient (MFCC), and Mel Spectrum. The results showed the highest accuracy of 91.76% for the Arabic-speakers model and 86.53% for the English-speakers model.