Spaces:
Sleeping
Sleeping
File size: 1,464 Bytes
88dc3ba d16271b 88dc3ba 335cb84 eb4d23a 88dc3ba c19f1d0 88dc3ba d16271b 88dc3ba 27f3b82 88dc3ba d16271b 27f3b82 88dc3ba 27f3b82 88dc3ba 7a0b87f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 |
import subprocess
import gradio as gr # Add this import statement
subprocess.run(["python", "-m", "pip", "install", "--upgrade", "pip"])
subprocess.run(["pip", "install", "gradio", "--upgrade"])
subprocess.run(["pip", "install", "datasets"])
subprocess.run(["pip", "install", "transformers"])
subprocess.run(["pip", "install", "librosa", "soundfile"])
subprocess.run(["pip", "install", "torch", "torchvision", "torchaudio", "-f", "https://download.pytorch.org/whl/torch_stable.html"])
from transformers import WhisperProcessor, WhisperForConditionalGeneration
from datasets import load_dataset
# Define the transcribe_audio function
def transcribe_audio(audio):
input_features = processor(audio, return_tensors="pt").input_features
predicted_ids = model.generate(input_features)
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
return transcription[0]
# load model and processor
processor = WhisperProcessor.from_pretrained("openai/whisper-small")
model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-small")
forced_decoder_ids = processor.get_decoder_prompt_ids(language="italian", task="transcribe")
# load dummy dataset and read audio files
ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
sample = ds[0]["audio"]
# Create Gradio interface
audio_input = gr.Audio()
gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
|