AI/ML Speech-to-Text Python Engineer

engineering

AI/ML Speech-to-Text Python Engineer

48 Commercial Plaza, Cavalry Ground Extension, Lahore

About IRG Global

IRG Global is a leading technology company specializing in software development, AI-powered solutions, digital transformation, and BPO services. With 8+ years of experience delivering high-performance tech solutions worldwide, IRG Global is now expanding its AI division and seeking top talent to build next-generation speech technologies.

About The Role

IRG Global is hiring an AI/ML Engineer with strong expertise in Speech-to-Text (STT) technologies and Python development. You will work on building advanced speech recognition systems, integrating cutting-edge models, and optimizing performance for real-time applications.

Key Responsibilities

• Build, train, and maintain Speech-to-Text (ASR) models using state-of-the-art frameworks.
• Develop Python-based STT pipelines and integrate them into IRG Global’s products and platforms.
• Fine-tune ASR models such as OpenAI Whisper, Wav2Vec2, DeepSpeech, or similar.
• Preprocess and manage audio datasets for training and evaluation.
• Optimize inference for low latency, high accuracy, and scalability.
• Deploy models using APIs, Docker, and cloud or on-prem infrastructure.
• Collaborate with product, development, and QA teams for end-to-end solution delivery.
• Research emerging speech technologies and propose improvements.
  
  

Required Skills & Qualifications

Bachelor’s or Master’s degree in Computer Science, AI, Data Science, or a related field.

Strong command of Python and ML frameworks (PyTorch, TensorFlow, NumPy, scikit-learn).

Experience Working With Speech-to-Text Frameworks

• OpenAI Whisper
• DeepSpeech
• Wav2Vec2 (Hugging Face)
• Or other ASR engines
• Understanding of audio processing, feature extraction, and signal processing.
• Experience working with ASR datasets (LibriSpeech, CommonVoice, etc.).
• Hands-on experience with model deployment (APIs, Docker, cloud/on-prem).
• Solid grasp of machine learning concepts, model tuning, and performance evaluation.
  
  

Preferred Skills

• Experience with real-time audio streaming and WebSocket integrations.
• Familiarity with NLP, text normalization, and acoustic modeling.
• Experience with GPU optimization or CUDA (plus point).
• Knowledge of LLMs and multimodal AI systems.
  
  

Soft Skills

• Strong analytical and problem-solving mindset.
• Ability to collaborate with cross-functional teams.
• Good communication skills and a proactive approach.
  
  

Why Join IRG Global?

• Work in an innovative, fast-growing AI environment.
• Opportunity to build transformative speech and AI technologies.
• Competitive salary, growth opportunities, and a collaborative work culture.
• Exposure to global clients and major industry projects.
  
  

Salary

PKR 80,000 to PKR 100,000

Apply to this job!