ya

Dockerfile

@@ -6,7 +6,6 @@ COPY app /code/app
 
 RUN pip install --upgrade pip
 RUN pip install -r /code/app/requirements.txt
-RUN python /code/app/nltk.py
 
 EXPOSE 7860
 

app/main.py

@@ -3,11 +3,9 @@ import nltk
 import uvicorn
 from fastapi import FastAPI, File, UploadFile
 from pydantic import BaseModel
-from sumy.parsers.plaintext import PlaintextParser
-from sumy.nlp.tokenizers import Tokenizer
-from sumy.summarizers.lsa import LsaSummarizer
 from groq import Groq
 
+
 GROQ_API_KEY = "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"
 groq_client = Groq(api_key=GROQ_API_KEY)
 
@@ -16,7 +14,7 @@ def save_to_file(content: str, filename: str) -> str:
         file.write(content)
     return filename
 
-def transcribe_and_summarize(audio_path: str):
+def transcribe_audio(audio_path: str):
     with open(audio_path, "rb") as audio_file:
         response = groq_client.audio.transcriptions.create(
             model="whisper-large-v3",
@@ -24,19 +22,7 @@ def transcribe_and_summarize(audio_path: str):
             response_format="text"
         )
     transcription = response
-
-    parser = PlaintextParser.from_string(transcription, Tokenizer("english"))
-    summarizer = LsaSummarizer()
-    summary_sentences = summarizer(parser.document, 5)
-    summarized_text = " ".join(str(s) for s in summary_sentences)
-
-    original_tokens = len(nltk.word_tokenize(transcription))
-    summarized_tokens = len(nltk.word_tokenize(summarized_text))
-    token_info = f"Asli: {original_tokens} token | Ringkasan: {summarized_tokens} token"
-
-    summarized_file = save_to_file(summarized_text, "summarized_transcription.txt")
-
-    return transcription, summarized_text, summarized_file, audio_path, token_info
+    return transcription
 
 def summarize_soap(dialogue: str):
     prompt_soap = f"""
@@ -97,7 +83,7 @@ async def full_process(audio: UploadFile = File(...)):
     with open(temp_audio_path, "wb") as f:
         f.write(await audio.read())
 
-    transcription, summarized_text, summarized_file, audio_path, token_info = transcribe_and_summarize(temp_audio_path)
+    transcription = transcribe_audio(temp_audio_path)
 
     soap_content, soap_file = generate_soap(transcription)
     tags_content, tags_file = generate_tags(transcription)
@@ -106,10 +92,7 @@ async def full_process(audio: UploadFile = File(...)):
 
     return {
         "transcription": transcription,
-        "summarized_text": summarized_text,
-        "summarized_file": summarized_file,
-        "audio_path": audio_path,
-        "token_info": token_info,
+        "audio_path": temp_audio_path,
         "soap_content": soap_content,
         "soap_file": soap_file,
         "tags_content": tags_content,

app/nltk.py

@@ -1 +0,0 @@
-import nltk