ML Practitioner working on Luhya Live Transcription Model
AudioTranscription
I built and trained a real-time transcription system for Luhya family names using labeled audio data. My workflow involved utilizing OpenAI’s Whisper model and organizing a domain-specific labeled dataset to fine-tune for speech recognition tasks. This required systematic transcription and validation to ensure language accuracy and dataset integrity.
• Processed and labeled audio samples focused on Luhya family names.
• Employed supervised learning methods and OpenAI Whisper for developing ASR models.
• Ensured precise label quality by reviewing and validating transcriptions.
• Developed and structured datasets for real-world applicability.
I built and trained a real-time transcription system for Luhya family names using labeled audio data. My workflow involved utilizing OpenAI’s Whisper model and organizing a domain-specific labeled dataset to fine-tune for speech recognition tasks. This required systematic transcription and validation to ensure language accuracy and dataset integrity. • Processed and labeled audio samples focused on Luhya family names. • Employed supervised learning methods and OpenAI Whisper for developing ASR models. • Ensured precise label quality by reviewing and validating transcriptions. • Developed and structured datasets for real-world applicability.
2024 - 2024