Welcome to my collection of Google Colab and Kaggle notebooks for various AI tools. This repository hosts free, easy-to-use notebooks that you can run directly in your browser or cloud environment.
| Notebook Name | Description | Link | Video Tutorial |
|---|---|---|---|
| LTX-2.5 22B Distilled - Dual GPU Video Generator (Kaggle) | Official LTX-2.5 22B Distilled with 8-step high-speed video generation & synchronized stereo audio. Optimized for Kaggle GPU T4 x2. | Open in Kaggle | Video Tutorial |
| HYPIR - 4K Image Upscaler & Restoration (Colab) | Single-step high-fidelity image restoration and 4K upscaling using diffusion-yielded score priors. Optimized for Colab Free T4 GPU. | Open in Colab | Video Tutorial |
| MOSS-TTS v1.5 - Standalone Foundation Model (Kaggle) | High-fidelity 48 kHz stereo TTS and zero-shot voice cloning. Optimized for Kaggle T4 x2 GPU. | Open in Kaggle | Video Tutorial |
| ScenA Audio - Expressive Speech Generator (Colab) | Zero-Shot Voice Cloning, Intent-Aware TTS, and Multi-Speaker Dialogue using the ScenA Audio model. Optimized for Colab Free T4 GPU. | Open in Colab | Video Tutorial |
| Krea 2 Turbo - Fast Text-to-Image Generator (Kaggle) | Ultra-fast text-to-image generation powered by the Krea 2 Turbo model. Optimized for Kaggle T4 GPU. | Open in Kaggle | Video Tutorial |
| LTX-2.3 22B MSR Ref Distilled 1.1 (Kaggle) | Multi-Reference Consistent Video Generation (Ingredients-to-Video) utilizing LTX-2.3 22B Distilled 1.1 + LiconStudio MSR LoRA. Optimized for Kaggle T4 GPU. | Open in Kaggle | Video Tutorial |
| Dots.TTS - Zero-Shot Autoregressive TTS | 2B Parameter Fully Continuous Autoregressive TTS Foundation Model. Features zero-shot speaker cloning & multiple checkpoints (Base, Soar, MF). Optimized for Colab Free T4 GPU. | Open in Colab | Video Tutorial |
| SeedVC - Voice Changer & Singing Cloner | Zero-Shot Voice Conversion · Vocal Demixing · Song Cover Cloning. Powered by SeedVC & RoFormer. Optimized for Colab Free T4 GPU. | Open in Colab | Video Tutorial |
| Scenema Audio Expressive Speech Generator (Kaggle) | Zero-Shot Voice Cloning, Intent-Aware TTS, and Multi-Speaker Dialogue using the Scenema Audio model. Optimized for Kaggle T4 GPU. | Open in Kaggle | Video Tutorial |
| LTX-2.3 22B Distilled 1.1 Q4 Text-to-Video (Kaggle) | Q4 Quantized version of LTX-2.3 22B Distilled 1.1 Text-to-Video. Optimized for Kaggle P100 GPU. | Open in Kaggle | Video Tutorial |
| LTX-2.3 22B Audio-to-Video Q4 (Kaggle) | Q4 Quantized version of LTX-2.3 Audio-to-Video generation. Optimized for Kaggle P100 GPU. | Open in Kaggle | Video Tutorial |
| ACE-Step 1.5 XL Turbo 4B (Kaggle) | High-quality AI music generation with the 4B XL Turbo model. Optimized for Kaggle T4 GPU. | Open in Kaggle | Video Tutorial |
| VoxCPM2 - Multilingual TTS | 2B Parameters · 30 Languages · 48kHz Output · Voice Design & Cloning. Optimized for Colab T4 GPU. | Open in Colab | Video Tutorial |
| OmniVoice - 600+ Language Zero-Shot TTS | Voice Cloning · Voice Design · Auto Voice · 600+ Languages. Powered by k2-fsa/OmniVoice. Optimized for Colab Free T4. | Open in Colab | Video Tutorial |
| Cohere Transcribe | State-of-the-art ASR · 14 Languages · Long-form Audio · 2B Parameter Conformer Model. Optimized for Colab Free T4. | Open in Colab | Video Tutorial |
| LTX-2.3 22B Distilled (Kaggle) | Powerful video generation on Kaggle P100 GPU. Featuring Wan2GP engine + mmgp Profile 4. | Open in Kaggle | Video Tutorial |
| SoulX FlashHead (Lite & Pro) | Audio-Driven AI Talking Head Generator. Features fast Lite & high-quality Pro models. | Open in Colab | Video Tutorial |
| ACE-Step 1.5 Music Generator (Custom UI) | Custom UI Edition for ACE-Step 1.5. Built-in examples & formatting. Optimized for Free T4. | Open in Colab | Video Tutorial |
| ACE-Step 1.5 Music Generator (LoRA Training) | LoRA Training Edition for ACE-Step 1.5. Full Gradio Interface & Model Management. | Open in Colab | Video Tutorial |
| MOSS-TTS 1.7B | Zero-Shot Voice Cloning & TTS. Optimized for Colab Free Tier. | Open in Colab | Video Tutorial |
| HeartMuLa 3B Music Generator | Free & Open Source AI Music Generation. BF16 Optimized for Colab Free Tier. | Open in Colab | Video Tutorial |
| Qwen3-TTS 1.7B | Advanced Text-to-Speech AI. Features Voice Cloning, Custom Voice & Voice Design. | Open in Colab | Video Tutorial |
- Click the "Open in Colab" or "Kaggle" badge next to the notebook you want to use.
- This will open the notebook in Google Colab or Kaggle.
- For Colab: Connect to a GPU runtime (Runtime -> Change runtime type -> T4 GPU).
- For Kaggle: Set Settings -> Accelerator -> GPU P100 x1 and turn on Internet.
- Run the cells in order.
Feel free to open issues or submit pull requests if you have suggestions or improvements for the notebooks.