Spaces:
Sleeping
Sleeping
File size: 19,015 Bytes
43ceeff |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 |
import openai
import json
import requests
import base64
import os
import tempfile
import asyncio
import edge_tts
import time
import hashlib
import shutil
from typing import List, Dict, Any, Optional
class VirtualInterviewer:
def __init__(self, api_key: str):
"""Initialize the virtual interviewer with the OpenAI API key."""
self.api_key = api_key
self.questions_asked = []
self.user_answers = []
self.conversation_history = []
self.ideal_answers = {}
self.question_audio_paths = {}
# Create audio directory
self.audio_dir = self._create_audio_directory()
# Clean up any existing audio files
self._cleanup_audio_files()
# Initialize OpenAI client
try:
self.client = openai.OpenAI(api_key=api_key)
except Exception as e:
raise Exception(f"Failed to initialize OpenAI client: {str(e)}")
def _create_audio_directory(self) -> str:
"""Create a directory to store audio files."""
audio_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "audio_files")
os.makedirs(audio_dir, exist_ok=True)
return audio_dir
def _cleanup_audio_files(self):
"""Delete all temporary audio files from previous sessions."""
try:
if os.path.exists(self.audio_dir):
# Delete all files in the directory
for filename in os.listdir(self.audio_dir):
file_path = os.path.join(self.audio_dir, filename)
if os.path.isfile(file_path):
os.remove(file_path)
print(f"Cleaned up audio files in {self.audio_dir}")
except Exception as e:
print(f"Error cleaning up audio files: {str(e)}")
def generate_interview_questions(
self,
job_description: str,
interview_type: str,
difficulty_level: str,
key_topics: str,
num_questions: int,
generate_ideal_answers: bool = True
) -> List[str]:
"""Generate interview questions based on the job description and other parameters."""
try:
# Construct the system prompt based on whether we want ideal answers or not
if generate_ideal_answers:
system_prompt = f"""You are an expert interviewer for {interview_type} interviews.
Generate {num_questions} {difficulty_level.lower()} difficulty interview questions for a {interview_type.lower()} interview based on the following job description:
Job Description:
{job_description}
Key Topics to Focus on:
{key_topics if key_topics else "No specific topics provided."}
Please provide the questions and ideal answers in the following JSON format:
{{
"questions": [
{{
"question": "Question 1",
"ideal_answer": "Ideal answer for question 1"
}},
...
]
}}
Make sure the questions are challenging but appropriate for the {difficulty_level.lower()} difficulty level.
The ideal answers should be comprehensive and demonstrate expertise in the subject matter.
"""
else:
system_prompt = f"""You are an expert interviewer for {interview_type} interviews.
Generate {num_questions} {difficulty_level.lower()} difficulty interview questions for a {interview_type.lower()} interview based on the following job description:
Job Description:
{job_description}
Key Topics to Focus on:
{key_topics if key_topics else "No specific topics provided."}
Please provide the questions in a numbered list format.
Make sure the questions are challenging but appropriate for the {difficulty_level.lower()} difficulty level.
"""
# Make the API call to generate questions
response = self.client.chat.completions.create(
model="gpt-4o",
messages=[
{"role": "system", "content": system_prompt},
{"role": "user", "content": f"Generate {num_questions} {interview_type.lower()} interview questions for a {difficulty_level.lower()} difficulty level."}
],
temperature=0.7,
max_tokens=2000
)
# Extract the response content
response_content = response.choices[0].message.content
# Process the response based on whether we're expecting JSON or a simple list
if generate_ideal_answers:
try:
# Try to parse as JSON
json_response = self._extract_json(response_content)
# Extract questions and ideal answers
questions = []
for item in json_response.get("questions", []):
question = item.get("question", "")
ideal_answer = item.get("ideal_answer", "")
if question:
questions.append(question)
if ideal_answer:
self.ideal_answers[question] = ideal_answer
# If we couldn't extract questions from JSON, fall back to parsing as text
if not questions:
questions = self._parse_questions(response_content, num_questions)
# Generate ideal answers separately
self._generate_ideal_answers(questions, job_description, interview_type, difficulty_level)
except Exception as e:
# If JSON parsing fails, fall back to text parsing
questions = self._parse_questions(response_content, num_questions)
# Generate ideal answers separately
self._generate_ideal_answers(questions, job_description, interview_type, difficulty_level)
else:
# Parse as simple text
questions = self._parse_questions(response_content, num_questions)
# Store the generated questions
self.questions_asked = questions
return questions
except Exception as e:
raise Exception(f"Failed to generate interview questions: {str(e)}")
def generate_question_audio(self, question: str, voice_type: str) -> str:
"""Generate audio for a question using edge-tts."""
try:
# Check if we already have audio for this question
if question in self.question_audio_paths and os.path.exists(self.question_audio_paths[question]):
print(f"Using existing audio for question: {question[:30]}...")
return self.question_audio_paths[question]
# Create a unique filename based on the question content and timestamp
question_hash = hashlib.md5(question.encode()).hexdigest()
timestamp = int(time.time())
filename = f"question_{question_hash}_{timestamp}.mp3"
output_path = os.path.join(self.audio_dir, filename)
# Map voice type to edge-tts voice
voice_mapping = {
"male_casual": "en-US-GuyNeural",
"male_formal": "en-US-ChristopherNeural",
"male_british": "en-GB-RyanNeural",
"female_casual": "en-US-JennyNeural",
"female_formal": "en-US-AriaNeural",
"female_british": "en-GB-SoniaNeural"
}
# Get the voice name from the mapping, default to female casual
voice = voice_mapping.get(voice_type, "en-US-JennyNeural")
# Generate audio using edge-tts
async def generate_audio():
communicate = edge_tts.Communicate(question, voice)
await communicate.save(output_path)
# Run the async function
asyncio.run(generate_audio())
print(f"Generated audio for question: {question[:30]}... at {output_path}")
# Store the audio path for this question
self.question_audio_paths[question] = output_path
return output_path
except Exception as e:
print(f"Error generating audio: {str(e)}")
return ""
def get_question_audio_path(self, question: str) -> str:
"""Get the audio path for a question."""
# Check if we have an audio path for this question
if question in self.question_audio_paths:
# Verify the file exists
if os.path.exists(self.question_audio_paths[question]):
return self.question_audio_paths[question]
else:
# File doesn't exist, remove from dictionary
del self.question_audio_paths[question]
return ""
return ""
def _extract_json(self, text: str) -> Dict[str, Any]:
"""Extract JSON from text."""
try:
# Try to parse the entire text as JSON
return json.loads(text)
except json.JSONDecodeError:
# If that fails, try to extract JSON from the text
import re
json_match = re.search(r'```json\n(.*?)\n```', text, re.DOTALL)
if json_match:
try:
return json.loads(json_match.group(1))
except json.JSONDecodeError:
pass
# Try to find JSON between curly braces
json_match = re.search(r'({.*})', text, re.DOTALL)
if json_match:
try:
return json.loads(json_match.group(1))
except json.JSONDecodeError:
pass
# If all else fails, return an empty dict
return {}
def _generate_ideal_answers(self, questions: List[str], job_description: str, interview_type: str, difficulty_level: str):
"""Generate ideal answers for the questions."""
try:
# Prepare the prompt for generating ideal answers
prompt = f"""You are an expert in {interview_type} interviews.
For each of the following interview questions, provide an ideal answer that would impress the interviewer.
The answers should be comprehensive, demonstrate expertise, and be appropriate for a {difficulty_level.lower()} difficulty level interview.
Job Description:
{job_description}
Questions:
{json.dumps(questions)}
Please provide the answers in the following JSON format:
{{
"answers": [
{{
"question": "Question 1",
"ideal_answer": "Ideal answer for question 1"
}},
...
]
}}
"""
# Make the API call to generate ideal answers
response = self.client.chat.completions.create(
model="gpt-4o",
messages=[
{"role": "system", "content": "You are an expert interviewer providing ideal answers to interview questions."},
{"role": "user", "content": prompt}
],
temperature=0.7,
max_tokens=2000
)
# Extract the response content
response_content = response.choices[0].message.content
try:
# Try to parse as JSON
json_response = self._extract_json(response_content)
# Extract ideal answers
for item in json_response.get("answers", []):
question = item.get("question", "")
ideal_answer = item.get("ideal_answer", "")
if question and ideal_answer:
# Find the matching question in our list
for q in questions:
if question.lower() in q.lower() or q.lower() in question.lower():
self.ideal_answers[q] = ideal_answer
break
except Exception as e:
# If batch processing fails, fall back to individual processing
for question in questions:
if question not in self.ideal_answers:
self.ideal_answers[question] = f"Unable to generate ideal answer: {str(e)}"
except Exception as e:
# Handle any errors in the overall ideal answer generation process
print(f"Error generating ideal answers: {str(e)}")
# Ensure all questions have a fallback ideal answer
for question in questions:
if question not in self.ideal_answers:
self.ideal_answers[question] = "Unable to generate ideal answer due to an error."
def _parse_questions(self, questions_text: str, expected_count: int) -> List[str]:
"""Parse the questions from the text response."""
lines = questions_text.strip().split('\n')
questions = []
for line in lines:
line = line.strip()
if line and (line[0].isdigit() or line.startswith('- ')):
# Remove numbering or bullet points
cleaned_line = line.lstrip('0123456789.- ').strip()
if cleaned_line:
questions.append(cleaned_line)
# If we couldn't parse the expected number of questions, try a simpler approach
if len(questions) != expected_count:
questions = [line.strip() for line in lines if line.strip()][:expected_count]
return questions[:expected_count] # Ensure we return exactly the expected number
def get_next_question(self, question_index: int) -> str:
"""Get the next question from the list of generated questions."""
if 0 <= question_index < len(self.questions_asked):
return self.questions_asked[question_index]
return "No more questions available."
def store_user_answer(self, question: str, answer: str):
"""Store the user's answer to a question."""
self.user_answers.append({"question": question, "answer": answer})
self.conversation_history.append({"role": "assistant", "content": question})
self.conversation_history.append({"role": "user", "content": answer})
def get_ideal_answer(self, question: str) -> str:
"""Get the ideal answer for a question."""
return self.ideal_answers.get(question, "No ideal answer available for this question.")
def score_interview(self, job_description: str, interview_type: str, difficulty_level: str) -> Dict[str, Any]:
"""Score the interview based on the user's answers."""
try:
# Prepare the data for scoring
questions_and_answers = []
for qa in self.user_answers:
question = qa["question"]
answer = qa["answer"]
ideal_answer = self.get_ideal_answer(question)
questions_and_answers.append({
"question": question,
"answer": answer,
"ideal_answer": ideal_answer
})
# Prepare the prompt for scoring
prompt = f"""You are an expert interviewer for {interview_type} interviews.
Score the following interview answers based on the job description and difficulty level.
Job Description:
{job_description}
Difficulty Level: {difficulty_level}
For each question and answer, provide:
1. A score from 0 to 5 (where 5 is excellent)
2. Feedback on the answer
3. Include the ideal answer for comparison. The ideal answer should be a comprehensive and detailed answer that would impress the interviewer with bullet points.
Questions and Answers:
{json.dumps(questions_and_answers)}
Please provide the scores in the following JSON format:
{{
"overall_score": 4.5,
"overall_feedback": "Overall feedback on the interview performance",
"individual_scores": [
{{
"question": "Question 1",
"answer": "User's answer to question 1",
"ideal_answer": "Ideal answer to question 1",
"score": 4,
"feedback": "Feedback on the answer to question 1"
}},
...
]
}}
"""
# Make the API call to score the interview
response = self.client.chat.completions.create(
model="gpt-4o",
messages=[
{"role": "system", "content": "You are an expert interviewer scoring interview answers."},
{"role": "user", "content": prompt}
],
temperature=0.3,
max_tokens=2000
)
# Extract the response content
response_content = response.choices[0].message.content
try:
# Try to parse as JSON
json_response = self._extract_json(response_content)
return json_response
except Exception as e:
# If JSON parsing fails, return an error
return {
"overall_score": 0,
"overall_feedback": f"Failed to score the interview: {str(e)}",
"individual_scores": []
}
except Exception as e:
# If scoring fails, return an error
return {
"overall_score": 0,
"overall_feedback": f"Failed to score the interview: {str(e)}",
"individual_scores": []
}
|