README.md · Rookiezz/whisper-small-th at main

metadata

datasets:
  - mozilla-foundation/common_voice_16_1
base_model:
  - openai/whisper-small
language:
  - th
metrics:
  - wer

Translation

from transformers import pipeline
import torch

MODEL_NAME = "Rookiezz/whisper-small-th"
lang = "th"

device = 0 if torch.cuda.is_available() else "CPU"

pipe = pipeline(
    task="automatic-speech-recognition",
    model=MODEL_NAME,
    chunk_length_s=448,
    device=device,
)

pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(
    language=lang,
    task="translate"
)
result = pipe("your_audio_file_path")
print(result["text"])