chunkformer-rnnt-small-libri-100h

ChunkFormer Model

<style> img { display: inline; } </style>

Usage

Install the package:

pip install chunkformer

from chunkformer import ChunkFormerModel

# Load the model
model = ChunkFormerModel.from_pretrained("khanhld/chunkFormer-rnnt-small-libri-100h")

# For long-form audio transcription
transcription = model.endless_decode(
    audio_path="path/to/your/audio.wav",
    chunk_size=64,
    left_context_size=128,
    right_context_size=128,
    return_timestamps=True
)
print(transcription)

# For batch processing
audio_files = ["audio1.wav", "audio2.wav", "audio3.wav"]
transcriptions = model.batch_decode(
    audio_paths=audio_files,
    chunk_size=64,
    left_context_size=128,
    right_context_size=128
)

Training

This model was trained using the ChunkFormer framework. For more details about the training process and to access the source code, please visit: https://github.com/khanld/chunkformer

Paper: https://arxiv.org/abs/2502.14673

Citation

If you use this work in your research, please cite:

@INPROCEEDINGS{10888640,
    author={Le, Khanh and Ho, Tuan Vu and Tran, Dung and Chau, Duc Thanh},
    booktitle={ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
    title={ChunkFormer: Masked Chunking Conformer For Long-Form Speech Transcription},
    year={2025},
    volume={},
    number={},
    pages={1-5},
    keywords={Scalability;Memory management;Graphics processing units;Signal processing;Performance gain;Hardware;Resource management;Speech processing;Standards;Context modeling;chunkformer;masked batch;long-form transcription},
    doi={10.1109/ICASSP49660.2025.10888640}}