Skip to content
View neoandmatrix's full-sized avatar
💭
thinking for .....
💭
thinking for .....

Highlights

  • Pro

Organizations

@p-society

Block or report neoandmatrix

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
neoandmatrix/README.md

Hi, I'm Tushar 👋

I build real-time voice AI for Indian languages: speech-to-text, text-to-speech and phone agents. I like going down to the GPU kernel level to make them fast on cheap hardware. Before that I worked as a backend engineer, mostly in Go, Python and TypeScript.

Voice AI

  • rumik-fast: I wrote my own inference engine for a 3B Hindi/English text-to-speech model.
    • It runs in real time on a 4 GB laptop GPU (RTX 3050): 59 → 95–98 tokens/s, with first sound in about 0.5 s.
    • Under the hood: int4 weights, hand-written CUDA int4 GEMV kernels, fused Triton decode, flash-decoding and streaming Mimi decoding.
    • The int4 weights are on Hugging Face: bringagi/rumik-oss-1-int4.
  • gemma-stt-hindi: a Hindi speech-to-text model built on Google's EmbeddingGemma 2 audio encoder, with a CTC head and a KenLM decoder.
    • It reaches 5.4% CER on Lahaja, on par with Qwen3-ASR-1.7B.
    • It was trained on free Kaggle T4 GPUs.
    • The model is on Hugging Face: bringagi/gemma-stt-hindi.

Backend and systems

  • rce-go: a remote code execution service written in Go.
  • http-from-scratch: HTTP/1.x implemented from scratch in Go, to learn how it works underneath.
  • anymemory: a service for creating and storing persistent memories.
  • terminal-todo: a small terminal todo app, in Go.

Now

  • Working on a Hindi/Hinglish end-of-turn detector, on-device Indic speech recognition, and a benchmarking harness for GPU kernels.
  • Looking for roles in voice AI, inference optimization or backend. Reach me at tusharannand@gmail.com.

Pinned Loading

  1. anymemory anymemory Public

    Create and store persistent memories.

    TypeScript 2

  2. gemma-stt-hindi gemma-stt-hindi Public

    Hindi speech-to-text on the EmbeddingGemma 2 audio encoder (CTC + KenLM), trained on free Kaggle GPUs

    Python

  3. rce-go rce-go Public

    This repository contains code for my Remote Code Execution built in golang.

    Go

  4. rumik-fast rumik-fast Public

    Real-time rumik-oss-1 TTS on a 4 GB laptop GPU: int4 CUDA/Triton kernels, streaming Mimi, web UI

    Python