What would you do? System Design: Architect and design end-to-end speech processing pipelines, from data acquisition to model deployment. Ensure systems are scalable, efficient, and maintainable. Advanced Modeling: Develop and implement advanced machine learning models for speech recognition, speaker diarization, and related tasks. Utilize state-of-the-art techniques such as deep learning, transfer learning, and ensemble methods. Research and Development: Conduct research to explore new methodologies and tools in the field of speech processing. Publish findings and present at industry conferences. Performance Optimization: Continuously monitor and optimize system performance, focusing on accuracy, latency, and resource utilization. Collaboration: Work closely with product management, data science, and software engineering teams to define project requirements and deliver innovative solutions. Customer Interaction: Engage with customers to understand their needs and provide tailored speech solutions. Assist in troubleshooting and optimizing deployed systems. Documentation and Standards: Establish and enforce best practices for code quality, documentation, and model management within the team. Required Skills 2+ years of experience in speech processing, machine learning, and model deployment. Demonstrated expertise in leading projects and teams. Technical skills: • Excellent knowledge in Python / Java programming. • In-depth knowledge of speech processing frameworks like, Wave2vec, Kaldi, HTK, DeepSpeech and Whisper. • Experience with NLP, STT, Speech to Speech LLMs and frameworks like Nvidia NEMO, PyAnnote. • Proficiency in Python and machine learning libraries such as TensorFlow, PyTorch, or Keras. • Experience with large-scale ASR systems, speaker recognition, and diarization algorithms. • Strong understanding of neural networks, sequence-to-sequence models, transformers and attention mechanisms. • Familiarity with NLP techniques and their integration with speech systems. • Expertise in deploying models on cloud platforms and optimizing for real-time applications. Good to have: • Experience with low-latency streaming ASR systems. • Knowledge of speech synthesis, STT (Speech-to-Text) and TTS (Text-to-Speech) systems. • Experience in multilingual and low-resource speech processing.
