m
mukesh_karnan

Mukesh K

@mukesh_karnan

Generative AI and Speech AI Engineer

Índia
Inglês, Tâmil
Algumas informações são exibidas no idioma inglês.
Sobre mim
I am a Generative AI and Speech AI Engineer with over 4+ years of experience building production-grade AI applications using LLMs, ASR, TTS, and RAG. I specialize in optimizing AI models for edge devices using ONNX Runtime and LiteRT, with expertise in offline Speech-to-Speech translation and on-device deployment. I am proficient in PyTorch, Whisper, and AWS.... Saiba mais

Habilidades

m
mukesh_karnan
Mukesh K
offline • 
Tempo médio de resposta: 1 hora

Conheça meus serviços

Sites Personalizados
I will build a modern responsive website with ai features and automation
A partir deUS$ 80 /projetoMais detalhes

Portfólio

Experiência profissional

IIT(Institute_of Information Technology)

Speech & TTS Engineer (ML / DL / Generative AI)

IIT(Institute of Information Technology) • Período integral

Jan 2024 - May 2026 • 2 yrs 4 mos

Built an end-to-end offline Speech-to-Speech translation system using Whisper ASR, Gemma LLM, FastSpeech2, and HiFi-GAN for multilingual speech processing. Developed RAG-based AI systems integrating embeddings, document retrieval, context management, and LLMs for knowledge-grounded responses. Worked on Agentic AI workflows using LLMs for reasoning, tool integration, task orchestration, and multi-step execution. Optimized ASR/TTS and Generative AI models for Android and embedded devices using ONNX Runtime, LiteRT, pruning, quantization, and graph optimization. Integrated Sherpa-ONNX and implemented asynchronous inference, reducing latency and runtime memory for efficient edge AI deployment. Built AI benchmarking and evaluation tools to measure latency, CPU/RAM usage, model accuracy, and performance across multiple devices.