Job Description:

As an LLM Engineer with a focus on speech, you will be responsible for designing, developing, and optimizing models that process both text and speech. You will work closely with cross-functional teams, including data scientists, speech scientists, and software engineers, to create end-to-end solutions for applications such as speech recognition, text-to-speech synthesis, voice assistants, and conversational AI systems.

Key Responsibilities:

• Design and implement innovative deep learning models for speech-to-text and text-to-speech applications.
• Work with state-of-the-art LLMs (e.g., GPT, BERT, etc.) to enhance the integration of speech recognition and language understanding.
• Develop algorithms for natural language processing (NLP) tasks in the context of audio and speech data.
• Collaborate with speech recognition engineers to improve accuracy, real-time performance, and the robustness of voice interaction systems.
• Implement speech synthesis systems for diverse language models, ensuring clarity, intonation, and natural flow of speech.
• Contribute to the development and optimization of multi-modal systems that combine speech with visual or textual information (e.g., lip-reading, visual speech recognition).
• Evaluate and benchmark performance on large-scale speech datasets and improve system efficiency.
• Implement state-of-the-art machine learning models in production-ready systems.
• Stay up to date with the latest research in LLMs, NLP, and speech technologies.

Required Skills and Qualifications:

• Bachelor's or Master’s degree in Computer Science, Electrical Engineering, Computational Linguistics, or a related field.
• Strong experience with deep learning frameworks such as TensorFlow, PyTorch, or similar.
• Expertise in speech processing (ASR, TTS, voice activity detection, etc.) and related models (e.g., wav2vec, Tacotron).
• Solid understanding of large language models (LLMs) like GPT-3, BERT, or T5.
• Experience in working with audio processing tools, libraries, and frameworks (e.g., Kaldi, HuggingFace, Librosa).
• Proficiency in programming languages such as Python, C++, or Java.
• Familiarity with cloud services and deployment platforms (e.g., AWS, GCP, Azure) for model deployment.
• Knowledge of optimization techniques for real-time systems, model compression, and deployment in edge devices is a plus.
• Excellent communication skills and the ability to work collaboratively in a fast-paced environment.

Preferred Qualifications:

• PhD in a related field (Natural Language Processing, Speech, Machine Learning, etc.).
• Experience with reinforcement learning or unsupervised learning in the context of speech models.
• Publications or contributions to research in speech technologies or NLP.
• Experience in multi-lingual or cross-lingual speech and language systems.
• Familiarity with conversational AI and chatbots.