OLMoASR-tiny

This is the tiny variant of OLMoASR, extracted from the original allenai/OLMoASR repository.

Model Details

• Model Size: Tiny
• Language: English
• License: Apache 2.0
• Task: Automatic Speech Recognition (ASR)

Files Included

• OLMoASR-tiny.en.pt - Model weights (~244 MB)
• config.json - Model configuration
• tokenizer.json - Tokenizer
• tokenizer_config.json - Tokenizer configuration
• vocab.json - Vocabulary
• merges.txt - BPE merges
• special_tokens_map.json - Special tokens mapping
• added_tokens.json - Additional tokens
• generation_config.json - Generation settings
• preprocessor_config.json - Audio preprocessing config
• model.safetensors - Model in safetensors format

Original Repository

This model is part of the OLMoASR family. For more information, please visit the original repository:

• allenai/OLMoASR

Usage

# Add usage instructions here based on OLMoASR documentation

Citation

If you use this model, please cite the original OLMoASR work.