Shaik Shahid A.

AI Data Annotation Specialist & Native Hindi & Telugu Linguist with extensive experience preparing, verifying, and refining datasets for Automatic Speech Recognition (ASR), NLP models, and AI dubbing pipelines. I help AI platforms and tech teams audit audio, speech, and multilingual text to ensure model training data and automated outputs are precise, contextually accurate, and production-ready. Core Expertise: * AI Transcription Review & QA: Auditing machine-generated transcripts against complex, multi-speaker audio to correct phonetic errors, dropped words, and colloquial nuances. * Audio Segmentation & Annotation: Setting exact acoustic segment boundaries, speaker diarization tags, and timestamped subtitle files (.srt). * Speech & Language Datasets: Native Hindi and bilingual (English/Telugu/Urdu) data labeling, translation evaluation, and prompt/response quality assessment. * Acoustic Quality Control: Identifying background noise, speaker overlap, accent variations, and audio degradation across diverse recording environments. Recent Upwork Experience Includes: * Native Hindi Audio Transcription Reviewer: Verified and edited spoken audio datasets for accuracy and phonetic fidelity. * LoFT 2.0 Segmentation Annotator: Conducted strict acoustic speech segmentation and multi-speaker tagging for long-term machine learning workflows. * Timestamped Subtitling & QA: Created synchronized .srt tracks and refined dialogue cadence for digital video formats. Available for ongoing projects and team workflows (30+ hours/week). Committed to strict turnaround times, thorough guideline compliance, and dependable data accuracy.