Nanthasit/sakthai-tts-model

---

license: apache-2.0 language: en pipeline_tag: text-to-speech tags: [tts, kokoro, gguf, cpu, speech, sakthai]

SakThai TTS Model

Part of the House of Sak

Kokoro-82M TTS in GGUF Q8_0. Lightweight speech on CPU.

141 MB | Q8_0 | Speech/TTS

Usage

Browser TTS (no setup)

Try the SakThai TTS Space: https://huggingface.co/spaces/Nanthasit/sakthai-tts

Python (Edge TTS)

import edge_tts async def speak(): tts = edge_tts.Communicate("Hello", "en-US-JennyNeural") await tts.save("output.mp3")

Specs

Detail	Value
Parameters	82M
Size	141 MB
Type	Speech/TTS

SakThai Model Family

Model	Size	Type	Downloads
1.5B-merged	934 MB	Tool-calling GGUF	942
0.5B-merged	380 MB	Lightweight GGUF	785
7B-merged	15 GB	Full-size model	534
7B-128K	15 GB	Extended context	324
Coder-1.5B	1.1 GB	Code GGUF	15
Embedding	80 MB	Search	28
Vision-7B	3.9 GB	Multimodal GGUF	—
TTS-Model	141 MB	Speech GGUF	—
Multilingual Embedding	80 MB	50+ languages	—

Full collection →

Benchmarks

Metric	Value
Model size	135 MB
Parameters	82M
Quantization	Q8_0
Speed	Real-time on CPU
Use case	Lightweight text-to-speech