Axiveri
/

NaijaVox-V1

Automatic Speech Recognition

nigerian-english

nigerian-pidgin

african-languages

Model card Files Files and versions

Ctrl+K

Ctrl+K

1 contributor

History: 30 commits

ememzyvisuals's picture

Update README.md

7e43411 verified 10 days ago

audio_samples
feat: add audio samples for model card 21 days ago
whisper-finetune
cleanup: remove old custom architecture files 21 days ago
.gitattributes

2.75 kB
Upload 2 files 21 days ago
README.md

14 kB
Update README.md 10 days ago
config.json

1.25 kB
feat: add merged standalone NaijaVox-V1 model (Whisper-large-v3 + LoRA epoch2) 21 days ago
generation_config.json

193 Bytes
fix: correct decoder_start_token_id (was missing, caused empty transcriptions) 20 days ago
model.safetensors

3.09 GB
xet

feat: add merged standalone NaijaVox-V1 model (Whisper-large-v3 + LoRA epoch2) 21 days ago
naijavox_logo.png

2.39 MB
xet

Upload 2 files 21 days ago
preprocessor_config.json

316 Bytes
fix: add missing preprocessor_config.json (was breaking pipeline() for all users) 19 days ago
processor_config.json

410 Bytes
feat: add merged standalone NaijaVox-V1 model (Whisper-large-v3 + LoRA epoch2) 21 days ago
tokenizer.json

3.93 MB
feat: add merged standalone NaijaVox-V1 model (Whisper-large-v3 + LoRA epoch2) 21 days ago
tokenizer_config.json

2.1 kB
fix: add missing tokenizer_config.json to repo root 19 days ago