Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Marathi
whisper
Generated from Trainer
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("simran14/stt-1")
model = AutoModelForSpeechSeq2Seq.from_pretrained("simran14/stt-1")Quick Links
mr stt 1st finetuning
This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:
- eval_loss: 0.5771
- eval_wer: 25.9487
- eval_runtime: 958.3173
- eval_samples_per_second: 1.637
- eval_steps_per_second: 0.206
- epoch: 0.6098
- step: 1000
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 2
- mixed_precision_training: Native AMP
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for simran14/stt-1
Base model
openai/whisper-small
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="simran14/stt-1")