Aller au contenu
ModelPrésentéExterne
Télécharger

Cey 25 Mms Tts Mozilla Nmz

Fine-tuned facebook/mms-tts-nmz on Mozilla Data Collective data for Nawdem (nmz).

Sourcehuggingface.co/Umbaji/cey-25-mms-tts-mozilla-nmz/tree/1c282585f819a31a0c82ad462d73e8d3060b57f2
TypeOtherModèle de basefacebook/mms-tts-nmzVersionRevision 1c28258

Source du modèle

Type
Other
Taille du fichier
139 MB
Modèle de base
facebook/mms-tts-nmz
Version
vRevision 1c28258
Mis à jour
21 sept. 2026
Téléchargements
5
Tâches
Audio
Source
Hugging Face

Description de la source

MMS-TTS Nawdem — Mozilla Fine-tuned

Fine-tuned facebook/mms-tts-nmz on Mozilla Data Collective data for Nawdem (nmz).

Training Statistics

MetricValue
Training samples14070
Validation samples2483
Best val mel-L14.0182
Dataset sourceMozilla Data Collective

Usage

from transformers import VitsModel, VitsTokenizer
import torch, torchaudio

model = VitsModel.from_pretrained("Umbaji/cey-25-mms-tts-mozilla-nmz")
tokenizer = VitsTokenizer.from_pretrained("Umbaji/cey-25-mms-tts-mozilla-nmz")

inputs = tokenizer("your text here", return_tensors="pt")
with torch.no_grad():
    waveform = model(**inputs).waveform[0]

torchaudio.save("out.wav", waveform.unsqueeze(0), model.config.sampling_rate)

Training

  • Loss: Mel-spectrogram L1
  • Optimizer: AdamW (lr=2e-4, betas=(0.8, 0.99))
  • Scheduler: ExponentialLR γ=0.999
  • Epochs: 1 | Effective batch: 16

Fine-tuned: 2026-06-12 — Mozilla Data Collective + Eyaa-Tom

Sources

1 source
Hugging FacePrincipalVersion 1c282585f819a31a0c82ad462d73e8d3060b57f2
huggingface.co/Umbaji/cey-25-mms-tts-mozilla-nmz/tree/1c282585f819a31a0c82ad462d73e8d3060b57f2

Vérifié 21 sept.

Artefacts du modèle

1 artefact

model.safetensors

safetensors · 139 MB · SHA-256 2d266a223cd4…9394 · Hugging Face

Télécharger

Extraits de sources

2 extraits
Readme21 sept.huggingface.co/Umbaji/cey-25-mms-tts-mozilla-nmz

--- language: - nmz tags: - text-to-speech - mms-tts - vits - nmz - west-africa - mozilla-data-collective - finetuned license: cc-by-nc-4.0 base_model: facebook/mms-tts-nmz pipeline_tag: text-to-speech --- # MMS-TTS Nawdem — Mozilla Fine-tuned Fine-tuned [facebook/mms-tts-nmz](https://huggingface.co/facebook/mms-tts-nmz) on Mozilla Data Collective data for **Nawdem** (`nmz`). ## Training Statistics | Metric | Value | |--------|-------| | Training samples | 14070 | | Validation samples | 2483 | | Best val mel-L1 | 4.0182 | | Dataset source | Mozilla Data Collective | ## Usage ```python from transformers import VitsModel, VitsTokenizer import torch, torchaudio model = VitsModel.from_pretrained("Umbaji/cey-25-mms-tts-mozilla-nmz") tokenizer = VitsTokenizer.from_pretrained("Umbaji/cey-25-mms-tts-mozilla-nmz") inputs = tokenizer("your text here", return_tensors="pt") with torch.no_grad(): waveform = model(**inputs).waveform[0] torchaudio.save("out.wav", waveform.unsqueeze(0), model.config.sampling_rate) ``` ## Training - **Loss**: Mel-spectrogram L1 - **Optimizer**: AdamW (lr=2e-4, betas=(0.8, 0.99)) - **Scheduler**: ExponentialLR γ=0.999 - **Epochs**: 1 | **Effective batch**: 16 *Fine-tuned: 2026-06-12 — Mozilla Data Collective + Eyaa-Tom*

Note21 sept.huggingface.co/Umbaji/cey-25-mms-tts-mozilla-nmz

Source context: 5 downloads · 0 likes · Pipeline text-to-speech · Repo Umbaji/cey-25-mms-tts-mozilla-nmz