Skip to content
View aynursusuz's full-sized avatar
👩‍💻
Working from home
👩‍💻
Working from home
30 followers · 83 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Shark

Block or report aynursusuz

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
aynursusuz/README.md
Aynur Susuz — AI Engineer, Speech AI & Text-to-Speech. Building the systems behind the voice.

Multilingual TTS Voice Cloning & Design Model Fine-Tuning Speech Evaluation

AI Engineer at Vyvo Labs · One year building text-to-speech systems.

Hugging Face   /   LinkedIn   /   Email

Synthetic speech

~4M English speech clips — Qwen3-TTS voice clones. Conversational speech synthesis.~3.98M English speech clips — EchoTTS voice clones. 4,000 reference speakers.
3.29M Japanese speech clips — Irodori TTS v2. 10,000 reference voices.~2.97M English speech clips with DNSMOS per-utterance quality estimates.

Also published: Irodori Japanese voice clones — 2.99M clips ↗

Text for speech synthesis

Dataset Language Text rows
Mandarin conversations Chinese ~5.8M
EchoTTS source text English 4M
DeepSeek Flash Japanese ~2.80M
DeepSeek Flash English ~2.70M
DeepSeek Pro English 1M

Sizes from dataset cards and Hub metadata · September 2026

Models & evaluation

Hugging Face organizations

Python · PyTorch · Transformers · CUDA

Pinned Loading

  1. UNITTS UNITTS Public

    One Python interface for 9 open-source text-to-speech models: Kokoro, Chatterbox, Qwen3-TTS, Supertonic, Fish Audio S2 Pro, Echo-TTS, NeuTTS, MOSS-TTS and Higgs TTS 3. Swap one name, get a new voice.

    Python 15 1

  2. micrograd micrograd Public

    A tiny scalar-valued autograd engine and neural network library (Karpathy course)

    Jupyter Notebook 12 1

  3. build-nanogpt build-nanogpt Public

    Building a GPT language model from scratch following Karpathy's nanogpt tutorial

    Python 4 1

  4. omnivoice-pipeline omnivoice-pipeline Public

    Python 3

  5. simple-shape-detection simple-shape-detection Public

    Simple geometric shape detection using OpenCV contour analysis

    Python 2

  6. audio-quality-pipeline audio-quality-pipeline Public

    Python 1