Upload folder using huggingface_hub

main.py

@@ -1,16 +1,21 @@
 import gradio as gr
-from transformers import pipeline
+import whisper
+import base64
 
+# Whisper 모델 로드
+model = whisper.load_model("base")
 
-whisper = pipeline("automatic-speech-recognition", model="openai/whisper-small")
-trans_en_ko = pipeline("translation", model="Helsinki-NLP/opus-mt-en-ko")
-trans_ko_en = pipeline("translation", model="Helsinki-NLP/opus-mt-ko-en")
+# 음성을 텍스트로 변환하는 함수
+def transcribe_audio(audio):
+    result = model.transcribe(audio)
+    return result["text"], base64.b64encode(result["text"].encode()).decode() #type: ignore
 
-def transcribe_and_translate(audio):
-    text = whisper(audio)['text'] #type: ignore
-    trans_text = trans_en_ko(text)['translation_text'] #type: ignore
+# Gradio 인터페이스
+demo = gr.Interface(fn=transcribe_audio,
+    inputs=gr.Audio(label = '오디오', sources="microphone", type="filepath"),
+    outputs=[gr.Textbox(label='결과'), gr.Textbox(label='암호화된 결과')],
+    title='이우진의 Speech to Text', 
+    description='AI 60%, 나 40%  (AI 코드 40분동안 고치다가 new chat으로 다시 짜니까 바로됨) (Interface는 내가만듬) (제작시간 1시간)')
 
-    return text, trans_text
-
-demo = gr.Interface(transcribe_and_translate, 'audio', ['text', 'text'])
-demo.launch()
+# 앱 실행
+demo.launch(share=True, auth=('경진씨반', '123456'))

requirements.txt

@@ -1,4 +1,5 @@
-torch
-torchaudio
-gradio
-transformers
+torch, torchaudio, transformers, gradio
+torch
+torchaudio
+gradio
+transformers