Responsibilities

• Program primarily in Java, Python, C++, including software engineering, algorithms and API design
• Collaborate with a small team of talented engineers to design and optimize code
• Extend and maintain a large-scale, carrier-grade automatic speech recognition system and related products
• Convey detailed documentation of all designs and procedures
• Actively develop enhancements to acoustic modeling mechanisms
• Researching and implementing state of the art features into our suite of products
• Improve performance of language models and decoder algorithms

Knowledge

• Extensive background in statistical probability theory, speech algorithms, HMM, GMM, FST, acoustic modeling and advanced machine learning techniques
• Experience in Statistical Language Modeling (SLM), Natural Language Processing (NLP) and/or Information Retrieval (IR) is highly desired
• Solid programming experience in Java, C/C++, Python, specifically runtime analysis, and design
• Experience in designing and running ASR experiments using HTK or KALDI a plus
• Able to solve real-world problems using cutting-edge ideas and independent research
• Able to read, adapt and implement ideas and algorithms from journals and papers
• Able to transfer technical knowledge to peers and management, as well as novices
• Knowledge of mainstream speech recognition software, DSP theory, feature extraction, scripting and parsing techniques