README.md

# 1step ASR-NL for Slurp dataset

This is a NeMo ASR model fine-tuned for ASR tasks. It was trained on [dataset name] and achieves [performance metrics]. This model is suitable for [use cases].

## Model Details

- **Model type**: NeMo ASR
- **Architecture**: Conformer CTC
- **Language**: English
- **Training data**: Slurp dataset
- **Performance metrics**: [Metrics]

## Usage

To use this model, you need to install the NeMo library:

```bash
pip install nemo_toolkit
```

### How to run

```python
import nemo.collections.asr as nemo_asr

# Step 1: Load the ASR model from Hugging Face
model_name = 'WhissleAI/1step_asr_nl_slurp'
asr_model = nemo_asr.models.EncDecCTCModel.from_pretrained(model_name)

# Step 2: Provide the path to your audio file
audio_file_path = '/path/to/your/audio_file.wav'

# Step 3: Transcribe the audio
transcription = asr_model.transcribe(paths2audio_files=[audio_file_path])
print(f'Transcription: {transcription[0]}')
```