Update app.py

app.py

@@ -1,10 +1,6 @@
 from flask import Flask, render_template, request, jsonify
 from qdrant_client import QdrantClient
 from qdrant_client import models
-import torch.nn.functional as F
-import torch
-from torch import Tensor
-from transformers import AutoTokenizer, AutoModel
 from qdrant_client.models import Batch, PointStruct
 from pickle import load, dump
 import numpy as np
@@ -12,7 +8,6 @@ import os, time, sys
 from datetime import datetime as dt
 from datetime import timedelta
 from datetime import timezone
-from faster_whisper import WhisperModel
 import io
 import requests
 
@@ -20,8 +15,8 @@ app = Flask(__name__)
 
 # Faster Whisper setup
 # model_size = 'small'
-beamsize = 2
-wmodel = WhisperModel("guillaumekln/faster-whisper-small", device="cpu", compute_type="int8")
+# beamsize = 2
+# wmodel = WhisperModel("guillaumekln/faster-whisper-small", device="cpu", compute_type="int8")
 
 # Initialize Qdrant Client and other required settings
 qdrant_api_key = os.environ.get("qdrant_api_key")
@@ -29,15 +24,7 @@ qdrant_url = os.environ.get("qdrant_url")
 
 client = QdrantClient(url=qdrant_url, port=443, api_key=qdrant_api_key, prefer_grpc=False)
 
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-def average_pool(last_hidden_states: Tensor,
-                 attention_mask: Tensor) -> Tensor:
-    last_hidden = last_hidden_states.masked_fill(~attention_mask[..., None].bool(), 0.0)
-    return last_hidden.sum(dim=1) / attention_mask.sum(dim=1)[..., None]
-
-tokenizer = AutoTokenizer.from_pretrained('intfloat/e5-base-v2')
-model = AutoModel.from_pretrained('intfloat/e5-base-v2').to(device)
+# device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 def e5embed(query):
   batch_dict = tokenizer(query, max_length=512, padding=True, truncation=True, return_tensors='pt')
@@ -134,26 +121,6 @@ def delete_joke():
     client.delete(collection_name=collection_name, points_selector=models.PointIdsList(points=[int(joke_id)],),)
     return jsonify({"deleted": True})
 
-@app.route("/whisper_transcribe", methods=["POST"])
-def whisper_transcribe():
-    if 'audio' not in request.files: return jsonify({'error': 'No file provided'}), 400
-
-    audio_file = request.files['audio']
-    allowed_extensions = {'mp3', 'wav', 'ogg', 'm4a'}
-    if not (audio_file and audio_file.filename.lower().split('.')[-1] in allowed_extensions): return jsonify({'error': 'Invalid file format'}), 400
-
-    print('Transcribing audio')
-    audio_bytes = audio_file.read()
-    audio_file = io.BytesIO(audio_bytes)
-
-    segments, info = wmodel.transcribe(audio_file, beam_size=beamsize) # beamsize is 2.
-    text = ''
-    starttime = time.time()
-    for segment in segments:
-        text += segment.text
-    print('Time to transcribe:', time.time() - starttime, 'seconds')
-
-    return jsonify({'transcription': text})
 
 
 if __name__ == "__main__":