---
language:
  - pt
tags:
  - text-to-speech
  - tts
  - voice-cloning
  - portuguese
  - pt-br
  - fine-tuning
license: other
license_name: mixed
license_details: Each model inherits the license of its base model — see the table below and each folder's README.
pipeline_tag: text-to-speech
base_model:
  - firstpixel/F5-TTS-pt-br
  - coqui/XTTS-v2
  - Qwen/Qwen3-TTS-12Hz-1.7B-Base
  - fishaudio/fish-speech-1.5
  - canopylabs/3b-es_it-ft-research_release
base_model_relation: finetune
---

# Brazilian Portuguese TTS Checkpoints — Paraíba Accent

Fine-tuning checkpoints of five Text-to-Speech frameworks for **Brazilian Portuguese (Paraíba accent)**, produced for the master's dissertation *Incorporando Regionalismos e Sotaques em Modelos de Síntese de Fala* (Thomaz Diniz, UFCG, 2026). All experiments share the same voice corpus (FALA_PB) and include checkpoints, evaluation audio, loss curves, and training reports.

**Every checkpoint in this repository is a fine-tune of its respective base model** (listed below and in the References section) — except `F5TTS/filtered_zero_experiment/`, which was trained from scratch as a control experiment.

## Models

| Folder | Base / starting checkpoint | Method | Size | Base license |
|---|---|---|---|---|
| [`F5TTS/`](F5TTS/README.md) | F5-TTS pt-br by **firstpixel** | Full fine-tuning (+ one from-scratch run) | ~21 GB | CC-BY-NC-4.0 |
| [`XTTS/`](XTTS/README.md) | XTTS v2 (Coqui) | GPT fine-tuning | ~37 GB | Coqui Public Model License |
| [`QwenTTS/`](QwenTTS/README.md) | Qwen3-TTS-12Hz-1.7B-Base | Full fine-tuning | ~13 GB | Apache-2.0 |
| [`FishSpeech/`](FishSpeech/README.md) | Fish Speech 1.5 | LoRA + merge | ~2.5 GB | CC-BY-NC-SA-4.0 |
| [`OrpheusTTS/`](OrpheusTTS/README.md) | Orpheus 3B (es_it) + **ArtooDtoo** PT-BR adapter | LoRA (Unsloth) | ~800 MB | Llama 3 / Apache-2.0 |

## Repository layout

Every model folder (or training folder, for F5TTS) follows the same pattern:

```
<Model>/
├── checkpoints          # .pt/.pth/.ckpt files or framework-format folders
├── configs & reports    # setting.json, REPORT.md, etc.
├── graficos/            # loss curves and metrics (PNG + CSV)
└── inferencias/         # evaluation audio (.wav) + phrases/manifests
```

- `F5TTS/` has one extra level for its 3 training runs: `all/`, `filtered/`, `filtered_zero_experiment/`
- Folder-format checkpoints: `QwenTTS/checkpoint-step-*/`, `FishSpeech/final_merged_llama/`, `OrpheusTTS/final_adapter/`

## Training data

Proprietary Brazilian Portuguese voice corpus with Paraíba accent (~41k segments, ~4 s average per sample). The dataset itself is **not** included in this repository — only the checkpoints, metrics, and generated evaluation audio.

## Usage

Each model's README states the framework and version used for training. In short: load the checkpoint with the official inference pipeline of the corresponding framework (F5-TTS, Coqui TTS, Qwen3-TTS, Fish Speech, or Orpheus/Unsloth).

## References — base models and starting checkpoints

The experiments started from the following models/checkpoints:

- **F5-TTS**: fine-tuned from the Brazilian Portuguese checkpoint by firstpixel — [firstpixel/F5-TTS-pt-br](https://huggingface.co/firstpixel/F5-TTS-pt-br) (`firstpixelptbr/model_last.pt`), built on [F5-TTS](https://github.com/SWivid/F5-TTS) (SWivid). The `filtered_zero_experiment` run was trained from scratch (no pretrained weights).
- **XTTS v2**: fine-tuned from [coqui/XTTS-v2](https://huggingface.co/coqui/XTTS-v2) (Coqui AI).
- **Qwen3-TTS**: fine-tuned from [Qwen/Qwen3-TTS-12Hz-1.7B-Base](https://huggingface.co/Qwen) (Alibaba/Qwen).
- **Fish Speech**: LoRA fine-tuned from [fishaudio/fish-speech-1.5](https://huggingface.co/fishaudio/fish-speech-1.5) (Fish Audio).
- **Orpheus TTS**: LoRA fine-tuned with [Unsloth](https://github.com/unslothai/unsloth) from [canopylabs/3b-es_it-ft-research_release](https://huggingface.co/canopylabs/3b-es_it-ft-research_release) (Canopy Labs), continuing from the PT-BR adapter [ArtooDtoo/Orpheus_PTBR_FT_Unsloth_Chk-1800](https://huggingface.co/ArtooDtoo/Orpheus_PTBR_FT_Unsloth_Chk-1800).

## Citation

If you use these checkpoints, please cite the master's dissertation:

**ABNT:**

> DINIZ, Thomaz. **Incorporando regionalismos e sotaques em modelos de síntese de fala**. 2026. Dissertação (Mestrado em Ciência da Computação) — Universidade Federal de Campina Grande, Campina Grande, 2026.

**BibTeX:**

```bibtex
@mastersthesis{diniz2026regionalismos,
  author  = {Diniz, Thomaz},
  title   = {Incorporando Regionalismos e Sotaques em Modelos de S{\'i}ntese de Fala},
  school  = {Universidade Federal de Campina Grande},
  address = {Campina Grande, Brazil},
  year    = {2026},
  type    = {Master's dissertation}
}
```