Audio Data Labeling & Transcription Specialist – Multimodal Audio/Video Annotation System
OtherAudioTranscription
Processed and labeled multimodal and multilingual audio datasets for speech-to-text model evaluation, focusing on noisy audio recordings. Applied transcription guidelines and performed validation checks to maintain high dataset reliability and accuracy. Reduced manual transcription workload by optimizing the annotation process and improving transcription efficiency.
• Achieved 78% accuracy on transcriptions of noisy audio data.
• Utilized Whisper, PyTorch, and WER/CER tools for evaluation and labeling.
• Detected and flagged inconsistencies to improve overall dataset quality.
• Enhanced workflow efficiency within the transcription pipeline.
Processed and labeled multimodal and multilingual audio datasets for speech-to-text model evaluation, focusing on noisy audio recordings. Applied transcription guidelines and performed validation checks to maintain high dataset reliability and accuracy. Reduced manual transcription workload by optimizing the annotation process and improving transcription efficiency. • Achieved 78% accuracy on transcriptions of noisy audio data. • Utilized Whisper, PyTorch, and WER/CER tools for evaluation and labeling. • Detected and flagged inconsistencies to improve overall dataset quality. • Enhanced workflow efficiency within the transcription pipeline.
2023 - 2024