Recognition Stage Research Articles

A BCI system for imagined Bengali speech recognition

Non-invasive brain-computer interface (BCI) can be used in imagined speech recognition to aid people with speech disorders and has been investigated in different languages. However, this technology is in a very early stage for Bengali speech recognition. Therefore, the aim of this research is to develop a BCI system that can recognize imagined Bengali speech. In this study, a non-invasive 14-channel electroencephalography (EEG) headset was employed to record the EEG signals from 30 subjects. The subjects were instructed to imagine 11 Bengali vowels ▪ and 10 digits ▪. Then, we extracted four statistical features from the recorded EEG signals. A coarse-to-fine level classification was performed to categorize the imagined EEG data by exploiting these features. The data was classified at coarse level to distinguish between a vowel and a digit. Each vowel and digit were then classified at the fine level. In our classification, random forest achieved the highest recognition accuracy of 84.28% at the coarse level and 76.13% at the fine level. In order to comprehend the role played by each brain lobe in recognizing imagined speech, we additionally examined each of the brain lobes. The frontal lobe demonstrated the highest accuracy, with a coarse level accuracy of 80.65% and a fine level accuracy of 66.31%. Moreover, the proposed model outperformed the reported literature in this field.

ANALISIS PREFERENSI KONSUMEN DALAM PEMBELIAN BERAS PREMIUM DI PASAR MODERN KOTA BATURAJA

This research has directions to find out the stages of the consumer decision-making process in purchasing premium rice, and how consumer preferences relate to rice attributes in purchasing premium rice in the Modern Market of Baturaja City. This research was conducted by survey method. Samples were taken using the accidental sampling method of 60 consumers of premium rice in the Modern Market, Baturaja City. Data processing was carried out in a quantitative descriptive manner, using conjoint analysis. The results of this study indicate the stages of the purchasing decision process through the stages of need recognition, information search, evaluation of alternatives and purchasing decisions. Consumer preferences related to rice attributes in purchasing premium rice at the Modern Market in Baturaja City consider that the most important attributes are fluffiness, shape, size and color.

Hybrid heuristic mechanism for occlusion aware facial expression recognition scheme using patch based adaptive CNN with attention mechanism

In computer vision, the recognition of expressions from partially occluded faces is one of the serious problems. By the prior recognizing techniques it can solve the issue by various assumptions. A benchmark-guided branch was proposed for detecting and eliminating the manipulated features from the occluded regions since the human visual system is proficient for eliminating the occlusion and the appropriate focal point was obtained on the non-occluded areas. In recent years deep learning has attained a great place in the recognition of facial reactions Still, the precision of facial expression is affected by the occlusion and large skew. In this research work, a deep structure-based occlusionaware facial expression recognition mechanism is introduced to provide superior recognition results. Firstly, the required image is taken from publically provided online sources and the gathered images are subjected to the face extraction method. The face extraction method is done via the ViolaJones method for the extraction of redundant patterns from the original images. Secondly, the extracted face features are given to the pattern recognition stage, where the Adaptive CNN with Attention Mechanism (ACNN-AM) is introduced. This mechanism automatically forms the occluded region of the face and the focal point was on the most discriminative un-occluded regions. Moreover, the hidden patterns in the Occlusion aware facial expressions are identified through the Hybrid Galactic Swarm Yellow Saddle Goatfish Optimization (HGSYSGO). Finally, the overall effectiveness of the developed occlusion aware facial expression recognition model is examined through the comparative analysis of different existing baseline recognition techniques.

Determination of the Predictive Labor Intensity of Manufacturing Parts on CNC Lathes Using the Feature Recognition

The problem of labor intensity operational forecasting of manufacturing parts on CNC lathes is considered. As a method, the recognition of structural and technological elements (features) of a part according to its 2D model was chosen to calculate the predictive labor intensity of machining. The processing of parts of the “sleeve” type on CNC lathes is considered. The feature recognition process is implemented in the form of an applied CAD/CAM system, which is fed with an electronic 2D model and a table of parameters of the manufactured surfaces. These parameters include accuracy and surface roughness requirements, the type of thread being cut, its pitch, the coordinates of the start and end points of the thread. At the output, a running program (NC) is formed for a CNC machine, designed in the form of standard cycles. Automatic assembly of NC becomes possible due to the use of technological templates for processing typical structural elements specially developed for this purpose. Technological templates are developed for all features typical for turning operation (end, open zone, semi-open zone, closed zone, and thread). For external and internal surfaces of the part technological templates are different, because various tools and strategies for constructing trajectories are used to process them. Formalization of closed area processing is of particular difficulty. Therefore various types of closed zones (undercut, groove, groove for thread exit, and face undercut) are classified at the stage of recognition. A separate technological template for their processing has been created for each type of closed areas. To determine the complexity, a standard simulation module NC is used. The results of model adaptation to current production are obtained.

Transductive semantic knowledge graph propagation for zero-shot learning

The domain shift is a very common phenomenon in Zero-shot Learning (ZSL), because the data distribution between the source and target domain might not match well in real scenarios. This paper focuses on dealing with the domain drift problem in ZSL based on the semantic knowledge graph propagation. Our method consists of visual feature extraction (VFE) module, semantic feature extraction module (SFE) and feature mapping module (FM). In VFE module, the high-level visual features are extracted under Convolutional Neural Network (CNN) framework, where unseen samples participate in model training and stay away from seen ones under the premise of distinguishability. SFE module relies on the Graph Convolutional Network (GCN) framework and focuses on the transmission of semantic embeddings. A modified message aggregation and transformation strategy is proposed to effectively relieve the information smoothing phenomenon caused by the increase of GCN layers. In the recognition stage, AutoEncoder framework achieves a two-way visual and semantic interaction to further minimize the domain drift phenomenon. Simulation results on ImageNet dataset have demonstrated the performance of the proposed method.

Speech enhancement approach for body-conducted unvoiced speech based on Taylor–Boltzmann machines trained DNN

Communication becomes effectual while the verbal communication signal reaches the destination with profound characteristics. An application based on human voice attracts the researcher's attention. It leads the researcher to proceed towards the speech conversion for low voice humans having murmured voices as their attitude. The researchers introduced a technique for improving the body-conducted unvoiced speech for silent speech communication using Taylor-Boltzmann machine-trained optimized Deep Neural Network (DNN). The projected model includes two main stages: the training stage as well as recognition stage. In the training stage, the following functionalities are undergone: (i) pre-processing, (ii) feature extraction, and (iii) deep learning-based speech enhancement. Initially, the collected input Non-Audible Murmur (NAM) speech signal is subjected to the pre-processing stage, wherein the improved wiener filtering is employed for suppressing the noise in the input signal. From the pre-processed speech signal, the most relevant characteristics such as “spectral skewness, spectral chroma, spectral centroid and improved Mel Frequency Cepstral Coefficients (MFCC)” are extracted, and they are fused together. Then, using these fused features, the Taylor series with optimized DNN and Boltzmann Machine (Taylor-DNN-BM) model is trained. The final enhanced speech is acquired from the Taylor- optimized DNN-BM. The recognition phase encapsulates the feature extraction and deep learning-based speech enhancement stage. The input speech data enters into the feature extraction phase, wherein the characteristics such as “spectral skewness, spectral chroma, spectral centroid and improved (MFCC)” are retrieved. After that, these features are fused and fed as input to the Taylor-optimized DNN-BM model. From Taylor- optimized DNN-BM, the enhanced speech signal is acquired. In DNN, the weight function is fine-tuned using the newly projected Self Improved Flower Pollination Algorithm (SI-FPA) algorithm. This SI-FPA model is the conceptual improvement of the Flower Pollination Algorithm (FPA). Finally, the projected model is compared with the existing models to validate its efficiency in terms of speech enhancement.

HIDR: A Hierarchical Independent Detection and Recognition strategy for underwater acoustic multi-target recognition

This paper proposed HIDR (Hierarchical Independent Detection and Recognition), a novel strategy for underwater acoustic multi-target recognition. HIDR adopts a hierarchical architecture instead of the single-step traditional target recognition task, which is a relatively challenging task especially for the multi-target recognition. HIDR would firstly distinguish target signals from non-target signals as a detection stage, then pinpoint the specific target from all possible types of target signals as a recognition stage. This hierarchical architecture cannot only sharply reduce the difficulty of feature extraction, but also benefit for the higher recognition accuracy. A universal feature extraction strategy is utilized to extract two independent feature sets for the two stages respectively, which include auditory parameters especially adaptive to detection/recognition stages, as well as other common time/frequency domain features. Experiments with real sea trial data are conducted to compare HIDR with existing high-quality multi-target recognition schemes. Experimental results show that the proposed strategy can reach up to 88.71% of recognition accuracy in four-class recognition task, which is more than 5.25% higher than that of mainstream methods.

Visualization of Seeking Identity on Transnational Children In “Be Prepared” Graphic Novel

Be Prepared is a graphic novel about Vera, the daughter of immigrants from Russia. In her storyline, Vera faces many cultural challenges when dealing with American and other Russian immigrant children. Vera attempts to find identity amidst the differences between the original culture of Russia and the American culture as the country she lived in. The situation between 2 cultures is a challenge for transnational children. This study aims to analyze and see the meaning represented in the graphic novel, especially how transnational children obtain identity and adapt to their environment. The research method used in this research is semiotics and literature review. The semiotic method is used to see the relationship between signs to get the meaning. The semiotic theories used in this research are based on Roland Barthes’s theories. Another method is the literature review which is conducted by overviewing the signs through several published literature to expand the meanings. The base theory used in the literature review is a cultural study that consists of ideas about acculturation, immigration, and cross-cultural theories. From this research, it can be seen that in the comparison of situations of 2 birthday parties, there are expectations in acculturation. The worship scene in the Russian Orthodox church represents immigrant families' coping mechanism for acculturation stress. The signs seen in the graphic novel are also seen in the acculturation stage of Vera's character to gain identity, which can be divided into three stages, namely, the stage of seeking recognition, the stage of imitating, and the stage of integration.
 Keywords: transnationalism; immigrant; acculturation; graphic novel; sequential arts.

Radial bias in face identification

Human vision in the periphery is most accurate for stimuli that point towards the fovea. This so-called radial bias has been linked with the organization and spatial selectivity of neurons at the lowest levels of the visual system, from retinal ganglion cells onwards. Despite evidence that the human visual system is radially biased, it is not yet known whether this bias persists at higher levels of processing, or whether high-level representations are invariant to this low-level orientation bias. We used the case of face identity recognition to address this question. The specialized high-level mechanisms that support efficient face recognition are highly dependent on horizontally oriented information, which convey the most useful identity cues in the fovea. We show that face selective mechanisms are more sensitive on the horizontal meridian (to the left and right of fixation) compared to the vertical meridian (above and below fixation), suggesting that the horizontal cues in the face are better extracted on the horizontal meridian, where they align with the radial bias. The results demonstrate that the radial bias is maintained at high-level recognition stages and emphasize the importance of accounting for the radial bias in future investigation of visual recognition processes in peripheral vision.

Motor problems in autism: Co-occurrence or feature?

Motor features of autism have long been acknowledged by clinicians, researchers, and community stakeholders. Current DSM-5 and ICD-11 guidelines allow clinicians to assign a co-occurring diagnosis of developmental [motor] coordination disorder (DCD) for autistic individuals with significant motor problems. DCD is characterized by poor motor proficiency with an onset of symptoms in early development. Studies have shown considerable overlap in the behavioral motor features observed in autism and DCD. However, others indicate that motor problems in autism and DCD may stem from different underlying sensorimotor mechanisms. Regardless of whether autism has a unique motor phenotype or an overlap with DCD, changes need to be made in the clinical pipeline to address motor problems in autism at the stages of recognition, assessment, diagnosis, and intervention. Consensus is needed to address unmet needs in research on the etiology of motor problems in autism and their overlap with DCD, to optimize clinical practice guidelines. The development of screening and assessment tools for motor problems that are valid and reliable for use with autistic individuals is essential, and an evidence-based clinical pipeline for motor problems in autism is urgently needed. WHAT THIS PAPER ADDS: Motor problems in autism are highly prevalent, yet underdiagnosed and poorly managed. An evidence-based clinical pipeline for motor problems in autism is urgently needed.

MuAt-Va: Multi-Attention and Video-Auxiliary Network for Device-Free Action Recognition

With the growing popularity of internet of things (IoT) systems, device-free action recognition begins to attract extensive attentions due to its friendly feasibility in broad applications such as human-computer interaction and smart elderly care. Considering abundant information in the vision modality, existing methods adopt the cross-model methods for performance enhancement. However, the dependency of synchronous multimodal data in the collection and recognition stage brings into the vision weaknesses such as sensitivity to occlusion and privacy invasion. In this paper, we integrate multi-attention structure and auxiliary video information into a novel end-to-end deep learning framework named MuAt-Va, in which video soft labels learned in advance are utilized to teach the multi-attention WiFi feature training process without vision information involved during test. Specifically, in order to enlarge the application scope and reduce the data cost, we beforehand acquire videos under a satisfactory condition only once, and then leverage teacher-student mechanism to guide the WiFi stream. Instead of straightforwardly concatenating multi-antenna channel state information (CSI) from homogeneous wireless signals as previous works, we design a CSI subcarrier-wise, temporal-wise and view-wise attention module to assign different weights on the basis of data characteristics for the sensing task. Our experiments with multiple subjects data in two scenes demonstrate that MuAt-Va can accurately recognize human actions with more superior performances.

Spatiotemporal phase slip patterns for visual evoked potentials, covert object naming tasks, and insight moments extracted from 256 channel EEG recordings.

Phase slips arise from state transitions of the coordinated activity of cortical neurons which can be extracted from the EEG data. The phase slip rates (PSRs) were studied from the high-density (256 channel) EEG data, sampled at 16.384 kHz, of five adult subjects during covert visual object naming tasks. Artifact-free data from 29 trials were averaged for each subject. The analysis was performed to look for phase slips in the theta (4-7 Hz), alpha (7-12 Hz), beta (12-30 Hz), and low gamma (30-49 Hz) bands. The phase was calculated with the Hilbert transform, then unwrapped and detrended to look for phase slip rates in a 1.0 ms wide stepping window with a step size of 0.06 ms. The spatiotemporal plots of the PSRs were made by using a montage layout of 256 equidistant electrode positions. The spatiotemporal profiles of EEG and PSRs during the stimulus and the first second of the post-stimulus period were examined in detail to study the visual evoked potentials and different stages of visual object recognition in the visual, language, and memory areas. It was found that the activity areas of PSRs were different as compared with EEG activity areas during the stimulus and post-stimulus periods. Different stages of the insight moments during the covert object naming tasks were examined from PSRs and it was found to be about 512 ± 21 ms for the 'Eureka' moment. Overall, these results indicate that information about the cortical phase transitions can be derived from the measured EEG data and can be used in a complementary fashion to study the cognitive behavior of the brain.

How Do Adults with Dyslexia Recognize Spoken Words? Evidence from Behavioral and EEG Data

ABSTRACT Purpose In adults with dyslexia (DYS), the persistent influence of phonological deficits on spoken language processing has mainly been examined in either perceptual tasks or those tapping complex cognitive operations. Much less attention is devoted to spoken word recognition per se. Our study aimed to fill this gap. Method Adults with and without dyslexia (for both groups: N = 30, mean age = 21 years, 50% female, 100% white European) performed an auditory lexical decision task. Performance and ERP were recorded. Results Reaction times showed a lexicality effect in both groups although they differed in ERP responses to stimulus lexicality. Skilled readers showed the typical amplitude enhancement for pseudowords compared to words in a late phase of N400 (414-581msec) whereas DYS showed the opposite pattern in an earlier phase of N400 (246-413msec). Both groups showed a stronger negativity during pseudowords processing in the late post-lexical stage (582-800msec). Conclusions ERP data showed subtle differences between the two populations during the lexical stage of word recognition despite their comparable behavioral outcomes. We hypothesized that a stronger reliance on intact semantic knowledge might contribute to the general enhanced and sustained ERP responses to words in DYS across different phases of lexical processing, although confirmation is needed.

Obstacle Detection Method Based on RSU and Vehicle Camera Fusion

Road obstacle detection is an important component of intelligent assisted driving technology. Existing obstacle detection methods ignore the important direction of generalized obstacle detection. This paper proposes an obstacle detection method based on the fusion of roadside units and vehicle mounted cameras and illustrates the feasibility of a combined monocular camera inertial measurement unit (IMU) and roadside unit (RSU) detection method. A generalized obstacle detection method based on vision IMU is combined with a roadside unit obstacle detection method based on a background difference method to achieve generalized obstacle classification while reducing the spatial complexity of the detection area. In the generalized obstacle recognition stage, a VIDAR (Vision-IMU based identification and ranging) -based generalized obstacle recognition method is proposed. The problem of the low accuracy of obstacle information acquisition in the driving environment where generalized obstacles exist is solved. For generalized obstacles that cannot be detected by the roadside unit, VIDAR obstacle detection is performed on the target generalized obstacles through the vehicle terminal camera, and the detection result information is transmitted to the roadside device terminal through the UDP (User Data Protocol) protocol to achieve obstacle recognition and pseudo-obstacle removal, thereby reducing the error recognition rate of generalized obstacles. In this paper, pseudo-obstacles, obstacles with a certain height less than the maximum passing height of the vehicle, and obstacles with a height greater than the maximum passing height of the vehicle are defined as generalized obstacles. Pseudo-obstacles refer to non-height objects that appear to be “patches” on the imaging interface obtained by visual sensors and obstacles with a height less than the maximum passing height of the vehicle. VIDAR is a vision-IMU-based detection and ranging method. IMU is used to obtain the distance and pose of the camera movement, and through the inverse perspective transformation, it can calculate the height of the object in the image. The VIDAR-based obstacle detection method, the roadside unit-based obstacle detection method, YOLOv5 (You Only Look Once version 5), and the method proposed in this paper were applied to outdoor comparison experiments. The results show that the accuracy of the method is improved by 2.3%, 17.4%, and 1.8%, respectively, compared with the other four methods. Compared with the roadside unit obstacle detection method, the speed of obstacle detection is improved by 1.1%. The experimental results show that the method can expand the detection range of road vehicles based on the vehicle obstacle detection method and can quickly and effectively eliminate false obstacle information on the road.

A super-resolution-based license plate recognition method for remote surveillance

With the continuous development of deep learning, neural networks have made great progress in license plate recognition (LPR). Nevertheless, there is still room to improve the performance of license plate recognition for low-resolution and relatively blurry images in remote surveillance scenarios. When it is difficult to enhance the recognition algorithm, we choose super-resolution (SR) to improve the quality of license plate images and thereby provide clearer input for the subsequent recognition stage. In this paper, we propose an automatic super-resolution license plate recognition (SRLPR) network which consists of four parts separately: license plate detection, character detection, single character super-resolution, and recognition. In the training stage, firstly, LP detection model needs to be trained alone and then its detection results will be used to successively train the three subsequent modules. During the test phase, for each input image, the network can get its LP number automatically. We also collect an applicable and challenging LPR dataset called SRLP, which is collected from real remote traffic surveillance. The experimental results demonstrate that our method achieves comprehensive quality of SR images and higher recognition accuracy compared with state-of-the-art methods. The SRLP dataset and the code for training and testing SRLPR network are available at https://pan.baidu.com/s/1vnhRa-c-dBj6jlfBZV5w4g.

Pattern of EEG voltage and oscillations under stimulation of Mozart’s music and white noise for visual learning process

The focus on behavioural data as a prominent indicator for evaluating the effect of stimulation on human cognitive function neglects the understanding of the body’s physiological responses. Another remain limitation is the physiological features usually analyzed from learning encode databases. Hence, the current research objectively investigates the correlation of EEG voltage and oscillations pattern from visual learning encoding and recognition stages under stimulation of various circumstances based on EEG patterns with behavioural data. The acquired EEG signal from visual assessment under no-audio and audio circumstances was processed using the Daubhecies wavelet function. The most affected EEG channels were identified according to the peak mean voltage from each electrode channel. The statistical features of mean, standard deviation, peak amplitude, and relative power were obtained from brain rhythms. The evaluation showed that the most affected EEG channel was at the Fp2 electrode with a voltage of about 0.1 V. It was related to the state of subjects that were required to identify the symbol and position of provided assessment. Most EEG features gained the highest value for recognition than the encoding stage, especially the relative power showed a statistically significant value with a p-value < 0.05. This revealed that brain oscillations were higher during recognition than encoding because the brain needed to work hard to retrieve the stored visual information. The behavioural data were parallel and in line with EEG rhythms pattern, where the highest brain oscillation led to better visual score performance. Throughout the evaluation, it was found that the EEG voltage and rhythms pattern during encoding were correlated with recognition and behavioural data.

Моделирование процесса распознавания символов в нормативных документах организации

Improving the quality of classification of different documents is a purpose of modeling the optical character recognition. Non-digital documents, such as scanned or photographed documents, are difficult to classify correctly in electronic document management systems. A decision was made to simulate the process of optical character recognition in the regulatory documents of the organization. There have been considered various methods of modeling the process. The structure of departments for the electronic document management system is given. Methods of implementing optical character recognition (OCR) are considered. The stages of the OCR system development are revealed: image processing, segmentation, recognition. The methods of image processing are analyzed. The main processes associated with image processing are disclosed: alignment, blurring, binarization, finding contours, removing extra lines. Comparison of image blur methods is made. Two stages of image binarization are defined: conversion of a color image into a gray image, binarization of a gray image. The Kenny operator is proposed as a second stage of binarization, which is used to detect the boundaries of the image. The last stage of image processing is the process of removing extra lines. Algorithms for dividing text areas into segments are considered. 3 stages of segmentation are identified: string segmentation, word segmentation, character segmentation. A segmentation algorithm is defined based on calculating the average brightness of image pixels to search for different intervals: line spacing, word spacing, character spacing. Available popular online OCR services as well as some popular desktop programs are considered. A connection has been found between an artificial neural network and optical object recognition. To implement the recognition stage, it is proposed to use an artificial neural network.

A Lightweight Traffic Lights Detection and Recognition Method for Mobile Platform

Traffic lights detection and recognition (TLDR) is one of the necessary abilities of multi-type intelligent mobile platforms such as drones. Although previous TLDR methods have strong robustness in their recognition results, the feasibility of deployment of these methods is limited by their large model size and high requirements of computing power. In this paper, a novel lightweight TLDR method is proposed to improve its feasibility to be deployed on mobile platforms. The proposed method is a two-stage approach. In the detection stage, a novel lightweight YOLOv5s model is constructed to locate and extract the region of interest (ROI). In the recognition stage, the HSV color space is employed along with an extended twin support vector machines (TWSVMs) model to achieve the recognition of multi-type traffic lights including the arrow shapes. The dataset, collected in naturalistic driving experiments with an instrument vehicle, is utilized to train, verify, and evaluate the proposed method. The results suggest that compared with the previous YOLOv5s-based TLDR methods, the model size of the proposed lightweight TLDR method is reduced by 73.3%, and the computing power consumption of it is reduced by 79.21%. Meanwhile, the satisfied reasoning speed and recognition robustness are also achieved. The feasibility of the proposed method to be deployed on mobile platforms is verified with the Nvidia Jetson NANO platform.

ЕND-TO-END SPEECH RECOGNITION SYSTEMS FOR AGGLUTINATIVE LANGUAGES

With the improvement of intelligent systems, speech recognition technologies are being widely integrated into various aspects of human life. Speech recognition is applied to smart assistants, smart home infrastructure, the call center applications of banks, information system components for impaired people, etc. But these facilities of information systems are available only for common languages, like English, Chinese, or Russian. For low-resource language, these opportunities for information technologies are still not implemented. Most modern speech recognition approaches are still not tested on agglutinative languages, especially for the languages of Turkic group like Kazakh, Tatar, and Turkish Languages. The HMM-GMM (Hidden Markov Models - Gaussian Mixture Models) model has been the most popular in the field of Automatic Speech Recognition (ASR) for a long time. Currently, neural networks are widely used in different fields of NLP, especially in automatic speech recognition. In an enormous number of works application of neural networks within different stages of automatic speech recognition makes the quality level of this systems much better. Integral speech recognition systems based on neural networks are investigated in the article. The paper proves that the Connectionist Temporal Classification (CTC) model works precisely for agglutinative languages. The author conducted an experiment with the LSHTM neural network using an encoder-decoder model, which is based on the attention-based models. The result of the experiment showed a Character Error Rate (CER) equal to 8.01% and a Word Error Rate (WER) equal to 17.91%. This result proves the possibility of getting a good ASR model without the use of the Language Model (LM).

Pemanfaatan Linkedin Sebagai Digital Marketing PT Vfirst Komunikasi Indonesia Dalam Membangun Brand Awareness di Indonesia

The digital era, provides opportunities for B2B companies to market their products through social media. Social media is one of the digital marketing strategies that are most chosen by business people because it is effective enough to connect with customers so that it can create brand awareness, generate leads, audience conversions, and more. This is the goal of PT. VFirst Communications Indonesia or commonly called VFirst, to implement digital marketing strategies. However, VFirst does not yet have a strong marketing team in doing digital marketing because all marketing activities are still being taken over and regulated by the central marketing team in India. VFirst in Indonesia entrusts Lead Generation to carry out digital marketing strategies by using LinkedIn to build brand awareness. This study was conducted to determine the use of LinkedIn as VFirst digital marketing in building brand awareness. The research method used is qualitative with in-depth interview data collection techniques. The results showed that the implementation of digital marketing through LinkedIn in building brand awareness in accordance with the main concepts used, with the selection of appropriate channels and interesting content as well as an approach to prospective customers carried out by the Lead Generation division succeeded in creating brand awareness from VFirst at the stage of recognition as a digital communication solution company.&#x0D; &#x0D; &#x0D;