whisper-medium-asm-16384

whisper-medium-asm-16384

This model is a 4.76% medium version of openai/whisper-medium optimized for Assamese language via vocabulary size reduction using the trimming method.
This trimmed model should perform similarly to the original model with only 16,384 tokens and a much smaller memory footprint. However, it may not perform well for other languages as tokens not commonly used in the selected languages were removed from the vocabulary.

Model Statistics

Mining Dataset Statistics

• Number of texts used for mining: 200,000 texts
• Dataset: lbourdois/fineweb-2-trimming

Usage

from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
import librosa

# Pipeline function
processor = AutoProcessor.from_pretrained("alphaedge-ai/whisper-medium-asm-16384")
pipe = pipeline(
    "automatic-speech-recognition",
    model="alphaedge-ai/whisper-medium-asm-16384",
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    generate_kwargs={"language": "assamese", "task": "transcribe"},
)

# Loading and resampling at 16 kHz (required by Whisper)
audio_array, sampling_rate = librosa.load(audio_path, sr=16000)

# Result
result = pipe(audio_array)
print("Transcription :", result["text"])

Citations

Whisper

@misc{radford2022whisper,
  doi = {10.48550/ARXIV.2212.04356},
  url = {https://arxiv.org/abs/2212.04356},
  author = {Radford, Alec and Kim, Jong Wook and Xu, Tao and Brockman, Greg and McLeavey, Christine and Sutskever, Ilya},
  title = {Robust Speech Recognition via Large-Scale Weak Supervision},
  publisher = {arXiv},
  year = {2022},
  copyright = {arXiv.org perpetual, non-exclusive license}
}

Trimming blog post

@misc{hf_blogpost_trimming,
      title={Introduction to Trimming}, 
      author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
      year={2026},
      url={https://huggingface.co/blog/lbourdois/introduction-to-trimming}, 
}