Skip to content
View suraj-geddam's full-sized avatar

Block or report suraj-geddam

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
suraj-geddam/README.md

Yo, I'm Suraj. My friends call me Su.

Currently: Founding Engineer at a stealth startup.

I spent three years at Google working on job scheduling and resource management infrastructure for an internal workflow engine. Before that, I did my master's in CS at Columbia, after two years at Samsung R&D mostly building apps for smart fridges (lol), and before that, my bachelor's in CS at IIT Bombay.

These days I mess around with local models a lot.

Recent work:

  • mlx-qwen3-tts-longform: long-form narration in one cloned voice with Qwen3-TTS on MLX, at about realtime on a base M3
  • longecho: the original; I'd initially done this for Echo-TTS. It runs on PyTorch with an NVIDIA GPU.

surajgeddam.com · LinkedIn

My profile picture is from when I used to need glasses. I've since had LASIK done, lol.

Pinned Loading

  1. mlx-qwen3-tts-longform mlx-qwen3-tts-longform Public

    Long-form narration with Qwen3-TTS on Apple Silicon (MLX): one consistent cloned voice, natural joins, about realtime

    Python

  2. longecho longecho Public

    Long-form TTS with Echo-TTS: arbitrary-length text in one consistent cloned voice, with a web UI and HTTP API that stream audio as it generates.

    Python 1 1