Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Shona
whisper
audio
multi-stage-finetuning
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("Code-Dev/whisper-small-shona-v3")
model = AutoModelForSpeechSeq2Seq.from_pretrained("Code-Dev/whisper-small-shona-v3", device_map="auto")Quick Links
🌾 Verdi AI OS: Whisper Small Shona ASR (Stage 1c Mastered Core)
This repository hosts a completely standalone, fully compiled model optimized natively for the Shona language (sn-ZW).
📊 Training Progression Metrics
- Stage 1a (FLEURS Baseline): Baseline WER dropped to 37.15%
- Stage 1b (Bible Chunk 1): Dropped WER to 29.39%
- Stage 1c (Bible Chunk 2 & 3 Increments): Achieved new floor of 25.52% WER.
- Downloads last month
- 556
Evaluation results
- Test WER on google/fleurs + manassehzw/shona-bible-bdsc-alignedself-reported25.520
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Code-Dev/whisper-small-shona-v3")