changes

Dockerfile

@@ -15,11 +15,13 @@ RUN apt-get update && apt-get install -y \
 # Install any needed packages specified in requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Make port 80 available to the world outside this container
-EXPOSE 80
+RUN python download_models.py
+
+# Make port 7860 available to the world outside this container
+EXPOSE 7860
 
 # Define environment variable
-ENV NAME World
+ENV TRANSFORMERS_CACHE=/tmp/.cache
 
 # Run app.py when the container launches
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

download_models.py

@@ -0,0 +1,19 @@
+import os
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
+
+# Create the models directory if it doesn't exist
+os.makedirs("./models", exist_ok=True)
+os.makedirs("./models/tokenizer", exist_ok=True)
+os.makedirs("./models/model", exist_ok=True)
+
+print("Downloading and saving tokenizer...")
+tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-base-960h")
+tokenizer.save_pretrained("./models/tokenizer")
+print("Tokenizer saved successfully.")
+
+print("Downloading and saving model...")
+model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
+model.save_pretrained("./models/model")
+print("Model saved successfully.")
+
+print("Download and save process completed.")

main.py

@@ -16,8 +16,10 @@ MultiPartParser.max_file_size = 200 * 1024 * 1024
 app = FastAPI()
 
 # Load Wav2Vec2 tokenizer and model
-tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-base-960h")
-model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
+tokenizer = Wav2Vec2Tokenizer.from_pretrained("./models/tokenizer")
+model = Wav2Vec2ForCTC.from_pretrained("./models/model")
+english_words = download_word_list()
+
 
 # Function to download English word list
 def download_word_list():
@@ -28,7 +30,6 @@ def download_word_list():
     print("Word list downloaded.")
     return words
 
-english_words = download_word_list()
 
 # Function to count correctly spelled words in text
 def count_spelled_words(text, word_list):
@@ -114,7 +115,7 @@ async def unscripted_root(audio_file: UploadFile):
     
     # Calculate pronunciation score
     fraction = correct / (incorrect + correct)
-    score = round(fraction * 10, 2)
+    score = round(fraction * 100, 2)
     print("Pronunciation score for", transcription, ":", score)
     
     print("Pronunciation scoring process complete.")