For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Sourabh Singh

Sourabh Singh

Project: AI-Powered Assistive Mobility Tool for Visually Impaired

INDIA flag
Delhi, India
$5.00/hrIntermediateOtherLabelboxCVAT

Key Skills

Software

Other
LabelboxLabelbox
CVATCVAT
Label StudioLabel Studio
RoboflowRoboflow
SuperAnnotateSuperAnnotate
V7 LabsV7 Labs
Google Cloud Vertex AIGoogle Cloud Vertex AI

Top Subject Matter

Assistive Mobility/Computer Vision
Fraud Detection/Finance
Sales Analytics/Audio Processing

Top Data Types

ImageImage
TextText
AudioAudio

Top Task Types

Object Detection
Classification
Transcription
Bounding Box
Segmentation
Data Collection
Computer Programming Coding

Freelancer Overview

I am a final-year B.Tech student from Delhi Technological University with a strong foundation in data analysis and AI-related workflows. I have hands-on experience working with different data types including text, image, and audio, and I am familiar with key data labeling techniques such as object detection, classification, and transcription. I have worked on projects like an AI Voice Analytics Pipeline using speech-to-text models and a Credit Card Fraud Detection system using machine learning, which enhanced my understanding of structured and unstructured data processing. I am detail-oriented and committed to delivering high-quality annotations with accuracy and consistency. I am comfortable using tools and platforms for data labeling, and I continuously focus on improving efficiency and quality. My analytical mindset, combined with technical skills in Python and data handling, allows me to contribute effectively to AI training and data annotation tasks.

IntermediateHindiEnglish

Labeling Experience

Project: AI Voice Analytics Pipeline

OtherAudioTranscription
Developed an AI pipeline for analyzing sales call audio data using speech-to-text models. Labeled segments for transcription and downstream sentiment and intent analysis. Integrated audio data labeling into broader analytics for sales decision support. • Applied Whisper and Hugging Face models for accurate audio transcription. • Performed intent detection and sentiment classification on call transcripts. • Estimated sales probability based on labeled conversational cues. • Employed Python and NLP techniques for end-to-end pipeline development.

Developed an AI pipeline for analyzing sales call audio data using speech-to-text models. Labeled segments for transcription and downstream sentiment and intent analysis. Integrated audio data labeling into broader analytics for sales decision support. • Applied Whisper and Hugging Face models for accurate audio transcription. • Performed intent detection and sentiment classification on call transcripts. • Estimated sales probability based on labeled conversational cues. • Employed Python and NLP techniques for end-to-end pipeline development.

Not specified

Project: Credit Card Fraud Risk Analytics

OtherTextClassification
Built a machine learning model to detect fraudulent credit card transactions using labeled data. Conducted extensive data preprocessing and feature engineering for optimal model performance. Applied classification algorithms to distinguish between fraudulent and legitimate transactions. • Addressed data imbalance to enhance model robustness. • Employed evaluation metrics such as ROC-AUC for performance measurement. • Used Python and scikit-learn throughout the development cycle. • Labeled features for analysis, training, and prediction phases.

Built a machine learning model to detect fraudulent credit card transactions using labeled data. Conducted extensive data preprocessing and feature engineering for optimal model performance. Applied classification algorithms to distinguish between fraudulent and legitimate transactions. • Addressed data imbalance to enhance model robustness. • Employed evaluation metrics such as ROC-AUC for performance measurement. • Used Python and scikit-learn throughout the development cycle. • Labeled features for analysis, training, and prediction phases.

Not specified

Project: AI-Powered Assistive Mobility Tool for Visually Impaired

OtherImageObject Detection
Developed a computer vision-based navigation system using YOLOv8 object detection for visually impaired users. The project involved processing real-time camera input to detect objects in the user's path. Converted detection results into actionable audio prompts for safe navigation. • Utilized YOLOv8 for object detection on live video streams. • Engineered a low-latency inference pipeline for responsive user feedback. • Evaluated accuracy and speed metrics to refine system reliability. • Translated model outputs into real-time voice alerts for end-users.

Developed a computer vision-based navigation system using YOLOv8 object detection for visually impaired users. The project involved processing real-time camera input to detect objects in the user's path. Converted detection results into actionable audio prompts for safe navigation. • Utilized YOLOv8 for object detection on live video streams. • Engineered a low-latency inference pipeline for responsive user feedback. • Evaluated accuracy and speed metrics to refine system reliability. • Translated model outputs into real-time voice alerts for end-users.

Not specified

Education

H

Hindon Public School

Class XII, Science

Class XII
2020 - 2021
H

Hindon Public School

Class X, General Studies

Class X
2018 - 2019

Work History

I

Infosys Springboard

Data Science Trainee

N/A
2025 - 2025
A

AddressofChoice Realty

Data Analyst Intern

Noida
2025 - 2025