Doqso's picture
Se ha añadido Torch como dependencia
43f775e
raw
history blame
No virus
626 Bytes
import torch
from transformers import pipeline
import numpy as np
import gradio as gr
pipe = pipeline(
"automatic-speech-recognition", model="openai/whisper-base"
)
def transcribe(stream, nuevo_fragmento):
sr, y = nuevo_fragmento
y = y.astype(np.float32)
y /= np.max(np.abs(y))
if stream is not None:
stream = np.concatenate([stream, y])
else:
stream = y
return stream, pipe({"sampling_rate": sr, "raw": stream})["text"]
demo = gr.Interface(
transcribe,
["state", gr.Audio(sources=["microphone"], streaming=True)],
["state", "text"],
live=True,
)
demo.launch()