metadata
datasets:
- aether-raid/SGdataset
metrics:
- wer
base_model:
- openai/whisper-large-v3-turbo
pipeline_tag: automatic-speech-recognition
Whisper Large V3 Turbo (WLV3t) trained on sgatc with
- Frozen Encoders (FE)
- Loud Normalization (LN)
- The following Augmentations (TSHLBT):
- T: time stretch
- S: seven band parametric EQ
- H: high pass
- L: low pass
- B: band pass
- T: tanh distortion