Chimaobim Nwachukwu

Key Skills

Software

Appen

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Polygon

Segmentation

Freelancer Overview

I am an experienced AI voice data specialist with a strong background in data annotation, speech dataset curation, and Natural Language Processing. My work at Scale AI involved annotating and validating over 10,000 hours of speech data for automatic speech recognition (ASR) training, with a focus on accent and dialect variations, particularly West African English. I have developed pronunciation lexicons, created speech normalization rules, and contributed to TTS voice selection and prosody tuning. My academic training in phonetics, linguistics, and advanced NLP models allows me to approach data labeling with attention to detail and a deep understanding of language patterns, ensuring high-quality, accurate training data for AI models.

Labeling Experience

DLT

AppenImageBounding BoxPolygon

SCOPE: Worked on large-scale speech and text annotation projects supporting the development and optimization of Automatic Speech Recognition (ASR), Text-to-Speech (TTS), and Natural Language Processing (NLP) systems. SPECIFIC DATA LABELLING TASKS PERFORMED: - Speech segmentation and timestamp alignment - Phoneme and pronunciation labeling - Speaker diarization (multi-speaker identification) - Accent and dialect annotation. PROJECT SIZE - Annotated and reviewed 5000–12,000+ hours of speech data across multiple projects - Processed 200,000+ text utterances for NLP training - Supported datasets covering 4+ accents and multiple English variants QUALITY MEASURES ADHERED TO: - Strict compliance with client annotation guidelines and style guides - Implemented multi-pass review and peer validation workflows. - Used structured feedback loops to continuously reduce labeling errors

SCOPE: Worked on large-scale speech and text annotation projects supporting the development and optimization of Automatic Speech Recognition (ASR), Text-to-Speech (TTS), and Natural Language Processing (NLP) systems. SPECIFIC DATA LABELLING TASKS PERFORMED: - Speech segmentation and timestamp alignment - Phoneme and pronunciation labeling - Speaker diarization (multi-speaker identification) - Accent and dialect annotation. PROJECT SIZE - Annotated and reviewed 5000–12,000+ hours of speech data across multiple projects - Processed 200,000+ text utterances for NLP training - Supported datasets covering 4+ accents and multiple English variants QUALITY MEASURES ADHERED TO: - Strict compliance with client annotation guidelines and style guides - Implemented multi-pass review and peer validation workflows. - Used structured feedback loops to continuously reduce labeling errors

2025

Education

Work History