# tiny-voca

> tiny-voca is a small English text-to-speech model with two voices, F2 (female) and M2 (male). The complete model, including the waveform generator, has 4.5M parameters and was trained from scratch on about 18 hours of synthetic speech. It runs about 70x faster than real time on four CPU threads.

- **Organisation:** n4ze3m
- **Category:** Text-to-Speech
- **Parameters:** 5M
- **Licence:** Apache 2.0 (Commercial use OK)
- **Released:** 2026-11-14
- **Hugging Face:** 9.3K downloads (30 days), 377 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q4_K_M | 0.5 |
| Q8_0 | 0.51 |
| F16 | 0.51 |

## Runs with

- [ONNX Runtime](https://onnxruntime.ai)

## About

tiny-voca is a small English text-to-speech model with two voices, F2 (female) and M2 (male). The complete model, including the waveform generator, has 4.5M parameters and was trained from scratch on about 18 hours of synthetic speech. It runs about 70x faster than real time on four CPU threads.

• Code: https://github.com/n4ze3m/tiny-voca
• Phonemizer: https://github.com/n4ze3m/dizione
• Dataset: https://huggingface.co/datasets/n4ze3m/tiny-voca

SHA256:

• tiny-voca.pt: ebb22871524c5732520fb1652e6aefcec708a7452e83bdd00adcf6181db5ef70
• onnx/model.onnx: fe6d918791a15b82e37465981d7f27a070e46b46635e5a3482fc8dff87a1ae46

## Alternatives

- [cloud0day3/antalia-mini](https://selfhost.directory/models/cloud0day3/antalia-mini) (8M, Apache 2.0, ≈0.52 GB): Türkçe özet. Antalia-2 Mini, tek sesli, küçük bir Türkçe metinden konuşmaya (TTS) modelidir: toplam 7,6 milyon parametre, 48 kHz çıkış, sıradan bir işlemcide gerçek zamandan çok daha hızlı.
- [owensong/Inflect-Nano-v1](https://selfhost.directory/models/owensong/inflect-nano-v1) (5M, Apache 2.0, ≈0.51 GB): Inflect-Nano-v1 is a tiny English text-to-speech model with 4.63M total inference parameters, including its vocoder.
- [canberkkkkkk/ema-lightning](https://selfhost.directory/models/canberkkkkkk/ema-lightning) (Apache 2.0): Tiny, fast and accurate Turkish text to speech. A 5.6M-parameter DiT acoustic model and a 3M-parameter vocoder: 8.6M in total, about 34 MB. Apache 2.0.
- [Supertone/supertonic-3](https://selfhost.directory/models/supertone/supertonic-3) (OpenRAIL): Supertonic is a lightweight text-to-speech system for local inference. It runs with ONNX Runtime entirely on your device, with no cloud call required for synthesis.
- [IndexTeam/IndexTTS-2.5](https://selfhost.directory/models/indexteam/indextts-2.5) (bilibili-model-license): IndexTTS-2.5 is a zero-shot text-to-speech model that clones a voice from a single reference audio clip. It supports Chinese, English, Japanese, Spanish and Arabic, with cross-lingual voice transfer and emotion control disentangled from timbre.
- [google/magenta-realtime-2](https://selfhost.directory/models/google/magenta-realtime-2) (CC BY 4.0): Magenta RealTime 2 is offered under a combination of licenses: the codebase is licensed under Apache 2.0, and the model weights under Creative Commons Attribution 4.0 International. In addition, we specify the following usage terms:

More: https://selfhost.directory/models/n4ze3m/tiny-voca/alternatives

## Runs well on

- [GeForce RTX 4060](https://selfhost.directory/models/n4ze3m/tiny-voca/on/rtx-4060): F16
- [Apple M4 (16 GB)](https://selfhost.directory/models/n4ze3m/tiny-voca/on/apple-m4-16gb): F16
- [Radeon RX 6700 XT](https://selfhost.directory/models/n4ze3m/tiny-voca/on/rx-6700-xt): F16
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/n4ze3m/tiny-voca/on/rtx-3060): F16
- [Arc B580](https://selfhost.directory/models/n4ze3m/tiny-voca/on/arc-b580): F16
- [Tesla T4](https://selfhost.directory/models/n4ze3m/tiny-voca/on/tesla-t4): F16

## Similar size

- [owensong/Inflect-Nano-v1](https://selfhost.directory/models/owensong/inflect-nano-v1) (5M)

## FAQ

**How much VRAM does tiny-voca need?**
About 0.5 GB at FP16 / BF16.

**Can tiny-voca run on the RTX 4060?**
Yes. It fits at FP16 / BF16 (about 0.5 GB). It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.

**Can I use tiny-voca commercially?**
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.

## Links

- huggingface: https://huggingface.co/n4ze3m/tiny-voca
- github: https://github.com/n4ze3m/tiny-voca

---
Source: https://selfhost.directory/models/n4ze3m/tiny-voca