devops/encoach_backend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add save endpoints but dont't actually save.

Browse Source
This commit is contained in:
Cristiano Ferreira
2023-11-18 23:18:46 +00:00
parent 73324909f6
commit 0bcf362b3f
3 changed files with 1333 additions and 687 deletions
Download Patch File Download Diff File Expand all files Collapse all files

370
app.py
View File

@@ -17,6 +17,7 @@ import re
from dotenv import load_dotenv
from heygen.AvatarEnum import AvatarEnum
from templates.question_templates import *
load_dotenv()
@@ -30,6 +31,7 @@ jwt = JWTManager(app)
cred = credentials.Certificate(os.getenv("GOOGLE_APPLICATION_CREDENTIALS"))
firebase_admin.initialize_app(cred)
@app.route('/healthcheck', methods=['GET'])
def healthcheck():
return {"healthy": True}
@@ -54,7 +56,8 @@ def get_listening_section_1_question():
print("Generated conversation: " + str(processed_conversation))
start_id = 1
exercises = generate_listening_conversation_exercises(unprocessed_conversation, req_exercises, number_of_exercises_q,
exercises = generate_listening_conversation_exercises(unprocessed_conversation, req_exercises,
number_of_exercises_q,
start_id)
return {
"exercises": exercises,
@@ -64,28 +67,6 @@ def get_listening_section_1_question():
return str(e)
@app.route('/save_listening_section_1', methods=['POST'])
@jwt_required()
def save_listening_section_1_question():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening1Template()
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
# TODO it's the conversation audio, still work to do on text-to-speech
text_to_speech(question["audio"]["conversation"], sound_file_path)
file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
question["audio"]["source"] = file_url
if save_to_db("listening", question):
return question
else:
raise Exception("Failed to save question: " + question)
except Exception as e:
return str(e)
@app.route('/listening_section_2', methods=['GET'])
@jwt_required()
def get_listening_section_2_question():
@@ -113,27 +94,6 @@ def get_listening_section_2_question():
return str(e)
@app.route('/save_listening_section_2', methods=['POST'])
@jwt_required()
def save_listening_section_2_question():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening2Template()
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
text_to_speech(question["audio"]["text"], sound_file_path)
file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
question["audio"]["source"] = file_url
if save_to_db("listening", question):
return question
else:
raise Exception("Failed to save question: " + question)
except Exception as e:
return str(e)
@app.route('/listening_section_3', methods=['GET'])
@jwt_required()
def get_listening_section_3_question():
@@ -153,7 +113,8 @@ def get_listening_section_3_question():
print("Generated conversation: " + str(processed_conversation))
start_id = 21
exercises = generate_listening_conversation_exercises(unprocessed_conversation, req_exercises, number_of_exercises_q,
exercises = generate_listening_conversation_exercises(unprocessed_conversation, req_exercises,
number_of_exercises_q,
start_id)
return {
"exercises": exercises,
@@ -163,27 +124,6 @@ def get_listening_section_3_question():
return str(e)
@app.route('/save_listening_section_3', methods=['POST'])
@jwt_required()
def save_listening_section_3_question():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening2Template()
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
text_to_speech(question["audio"]["text"], sound_file_path)
file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
question["audio"]["source"] = file_url
if save_to_db("listening", question):
return question
else:
raise Exception("Failed to save question: " + question)
except Exception as e:
return str(e)
@app.route('/listening_section_4', methods=['GET'])
@jwt_required()
def get_listening_section_4_question():
@@ -211,23 +151,30 @@ def get_listening_section_4_question():
return str(e)
@app.route('/save_listening_section_4', methods=['POST'])
@app.route('/listening', methods=['POST'])
@jwt_required()
def save_listening_section_4_question():
def save_listening():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening2Template()
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
text_to_speech(question["audio"]["text"], sound_file_path)
file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
question["audio"]["source"] = file_url
if save_to_db("listening", question):
return question
else:
raise Exception("Failed to save question: " + question)
data = request.get_json()
parts = data.get('parts')
template = getListeningTemplate()
for i, part in enumerate(parts, start=0):
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
if "conversation" in part["text"]:
conversation_text_to_speech(part["text"]["conversation"], sound_file_path)
else:
text_to_speech(part["text"], sound_file_path)
# file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
file_url = "test_url/" + firebase_file_path
template["parts"][i]["audio"]["source"] = file_url
template["parts"][i]["exercises"].append(part["exercises"])
# if save_to_db("listening", template):
# return template
# else:
# raise Exception("Failed to save question: " + parts)
return template
except Exception as e:
return str(e)
@@ -278,22 +225,6 @@ def get_writing_task_1_general_question():
return str(e)
@app.route('/save_writing_task_1', methods=['POST'])
@jwt_required()
def save_writing_task_1_question():
try:
# data = request.get_json()
# question = data.get('question')
# TODO ADD SAVE IMAGE TO DB
question = getListening2Template()
if save_to_db("writing", question):
return question
else:
raise Exception("Failed to save question: " + question)
except Exception as e:
return str(e)
@app.route('/writing_task2', methods=['POST'])
@jwt_required()
def grade_writing_task_2():
@@ -341,17 +272,20 @@ def get_writing_task_2_general_question():
return str(e)
@app.route('/save_writing_task_2', methods=['POST'])
@app.route('/writing', methods=['POST'])
@jwt_required()
def save_writing_task_2_question():
def save_writing_task():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening2Template()
if save_to_db("writing", question):
return question
else:
raise Exception("Failed to save question: " + question)
data = request.get_json()
exercises = data.get('exercises')
template = getWritingTemplate()
for i, exercise in enumerate(exercises, start=0):
template["exercises"][i]["prompt"] = exercise
# if save_to_db("writing", template):
# return template
# else:
# raise Exception("Failed to save writing: " + template)
return template
except Exception as e:
return str(e)
@@ -408,54 +342,6 @@ def get_speaking_task_1_question():
return str(e)
@app.route('/save_speaking_task_1', methods=['POST'])
@jwt_required()
def save_speaking_task_1_question():
try:
# data = request.get_json()
# question = data.get('question')
questions_json = getSpeaking1Template()
questions = []
for question in questions_json["questions"]:
result = create_video(question)
if result is not None:
sound_file_path = VIDEO_FILES_PATH + result
firebase_file_path = FIREBASE_SPEAKING_VIDEO_FILES_PATH + result
url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
video = {
"text": question,
"video_path": firebase_file_path,
"video_url": url
}
questions.append(video)
else:
print("Failed to create video for question: " + question)
if len(questions) == len(questions_json["questions"]):
speaking_pt1_to_insert = {
"exercises": [
{
"id": str(uuid.uuid4()),
"prompts": questions,
"text": "Listen carefully and respond.",
"title": questions_json["topic"],
"type": "speakingPart1"
}
],
"isDiagnostic": True,
"minTimer": 5,
"module": "speaking"
}
if save_to_db("speaking", speaking_pt1_to_insert):
return speaking_pt1_to_insert
else:
raise Exception("Failed to save question: " + speaking_pt1_to_insert)
else:
raise Exception("Array sizes do not match. Video uploading failing is probably the cause.")
except Exception as e:
return str(e)
@app.route('/speaking_task_2', methods=['POST'])
@jwt_required()
def grade_speaking_task_2():
@@ -509,54 +395,6 @@ def get_speaking_task_2_question():
return str(e)
@app.route('/save_speaking_task_2', methods=['POST'])
@jwt_required()
def save_speaking_task_2_question():
try:
# data = request.get_json()
# question = data.get('question')
questions_json = getSpeaking2Template()
questions = []
for question in questions_json["questions"]:
result = create_video(question)
if result is not None:
sound_file_path = VIDEO_FILES_PATH + result
firebase_file_path = FIREBASE_SPEAKING_VIDEO_FILES_PATH + result
url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
video = {
"text": question,
"video_path": firebase_file_path,
"video_url": url
}
questions.append(video)
else:
print("Failed to create video for question: " + question)
if len(questions) == len(questions_json["questions"]):
speaking_pt2_to_insert = {
"exercises": [
{
"id": str(uuid.uuid4()),
"prompts": questions,
"text": "Listen carefully and respond.",
"title": questions_json["topic"],
"type": "speakingPart2"
}
],
"isDiagnostic": True,
"minTimer": 5,
"module": "speaking"
}
if save_to_db("speaking", speaking_pt2_to_insert):
return speaking_pt2_to_insert
else:
raise Exception("Failed to save question: " + str(speaking_pt2_to_insert))
else:
raise Exception("Array sizes do not match. Video uploading failing is probably the cause.")
except Exception as e:
return str(e)
@app.route('/speaking_task_3', methods=['GET'])
@jwt_required()
def get_speaking_task_3_question():
@@ -612,50 +450,76 @@ def grade_speaking_task_3():
return str(e), 400
@app.route('/save_speaking_task_3', methods=['POST'])
@app.route('/speaking', methods=['POST'])
@jwt_required()
def save_speaking_task_3_question():
def save_speaking():
try:
# data = request.get_json()
# question = data.get('question')
questions_json = getSpeaking3Template()
questions = []
for question in questions_json["questions"]:
result = create_video(question)
data = request.get_json()
exercises = data.get('exercises')
template = getSpeakingTemplate()
# Speaking 1
# sp1_result = create_video(exercises[0]["question"], random.choice(list(AvatarEnum)))
sp1_result = "speaking_1"
if sp1_result is not None:
sound_file_path = VIDEO_FILES_PATH + sp1_result
firebase_file_path = FIREBASE_SPEAKING_VIDEO_FILES_PATH + sp1_result
# url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
url = "speaking_1_url"
sp1_video_path = firebase_file_path
sp1_video_url = url
template["exercises"][0]["text"] = exercises[0]["question"]
template["exercises"][0]["title"] = exercises[0]["topic"]
template["exercises"][0]["video_url"] = sp1_video_url
template["exercises"][0]["video_path"] = sp1_video_path
else:
print("Failed to create video for part 1 question: " + exercises[0]["question"])
# Speaking 2
# sp2_result = create_video(exercises[1]["question"], random.choice(list(AvatarEnum)))
sp2_result = "speaking_2"
if sp2_result is not None:
sound_file_path = VIDEO_FILES_PATH + sp2_result
firebase_file_path = FIREBASE_SPEAKING_VIDEO_FILES_PATH + sp2_result
# url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
url = "speaking_2_url"
sp2_video_path = firebase_file_path
sp2_video_url = url
template["exercises"][1]["prompts"] = exercises[1]["prompts"]
template["exercises"][1]["text"] = exercises[1]["question"]
template["exercises"][1]["title"] = exercises[1]["topic"]
template["exercises"][1]["video_url"] = sp2_video_url
template["exercises"][1]["video_path"] = sp2_video_path
else:
print("Failed to create video for part 2 question: " + exercises[1]["question"])
# Speaking 3
sp3_questions = []
avatar = random.choice(list(AvatarEnum))
for question in exercises[2]["questions"]:
# result = create_video(question, avatar)
result = "speaking_3"
if result is not None:
sound_file_path = VIDEO_FILES_PATH + result
firebase_file_path = FIREBASE_SPEAKING_VIDEO_FILES_PATH + result
url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
# url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
url = "speaking_3_url"
video = {
"text": question,
"video_path": firebase_file_path,
"video_url": url
}
questions.append(video)
sp3_questions.append(video)
else:
print("Failed to create video for question: " + question)
print("Failed to create video for part 3 question: " + question)
template["exercises"][2]["prompts"] = sp3_questions
template["exercises"][2]["title"] = exercises[2]["topic"]
if len(questions) == len(questions_json["questions"]):
speaking_pt3_to_insert = {
"exercises": [
{
"id": str(uuid.uuid4()),
"prompts": questions,
"text": "Listen carefully and respond.",
"title": questions_json["topic"],
"type": "speakingPart3"
}
],
"isDiagnostic": True,
"minTimer": 5,
"module": "speaking"
}
if save_to_db("speaking", speaking_pt3_to_insert):
return speaking_pt3_to_insert
else:
raise Exception("Failed to save question: " + str(speaking_pt3_to_insert))
else:
raise Exception("Array sizes do not match. Video uploading failing is probably the cause.")
# if save_to_db("speaking", template):
# return template
# else:
# raise Exception("Failed to save speaking: " + template)
return template
except Exception as e:
return str(e)
@@ -688,28 +552,6 @@ def get_reading_passage_1_question():
return str(e)
@app.route('/reading_passage_1', methods=['POST'])
@jwt_required()
def save_reading_passage_1_question():
try:
# data = request.get_json()
# question = data.get('question')
question = getListening1Template()
file_name = str(uuid.uuid4()) + ".mp3"
sound_file_path = AUDIO_FILES_PATH + file_name
firebase_file_path = FIREBASE_LISTENING_AUDIO_FILES_PATH + file_name
# TODO it's the conversation audio, still work to do on text-to-speech
text_to_speech(question["audio"]["conversation"], sound_file_path)
file_url = upload_file_firebase_get_url(FIREBASE_BUCKET, firebase_file_path, sound_file_path)
question["audio"]["source"] = file_url
if save_to_db("listening", question):
return question
else:
raise Exception("Failed to save question: " + question)
except Exception as e:
return str(e)
@app.route('/reading_passage_2', methods=['GET'])
@jwt_required()
def get_reading_passage_2_question():
@@ -765,6 +607,22 @@ def get_reading_passage_3_question():
except Exception as e:
return str(e)
@app.route('/reading', methods=['POST'])
@jwt_required()
def save_reading_passage():
try:
data = request.get_json()
parts = data.get('parts')
template = getReadingTemplate()
template["parts"] = parts
# if save_to_db("reading", template):
# return template
# else:
# raise Exception("Failed to save reading: " + template)
return template
except Exception as e:
return str(e)
@app.route('/level', methods=['GET'])
@jwt_required()

53
helper/speech_to_text_helper.py
View File

@@ -24,18 +24,19 @@ def text_to_speech(text: str, file_name: str):
aws_access_key_id=os.getenv("AWS_ACCESS_KEY_ID"),
aws_secret_access_key=os.getenv("AWS_SECRET_ACCESS_KEY")
)
voice = random.choice(ALL_NEURAL_VOICES)['Id']
# Initialize an empty list to store audio segments
audio_segments = []
tts_response = client.synthesize_speech(
Engine="neural",
Text=text,
OutputFormat="mp3",
VoiceId=random.choice(ALL_NEURAL_VOICES)['Id']
)
audio_segments.append(tts_response['AudioStream'].read())
for part in divide_text(text):
tts_response = client.synthesize_speech(
Engine="neural",
Text=part,
OutputFormat="mp3",
VoiceId=voice
)
audio_segments.append(tts_response['AudioStream'].read())
# Combine the audio segments into a single audio file
combined_audio = b"".join(audio_segments)
file_name = file_name + ".mp3"
# Save the combined audio to a single file
with open(file_name, "wb") as f:
f.write(combined_audio)
@@ -43,20 +44,6 @@ def text_to_speech(text: str, file_name: str):
print("Speech segments saved to " + file_name)
def conversation_text_to_speech(conversation: list, file_name: str):
# Create a dictionary to store the mapping of 'name' to 'voice'
name_to_voice = {}
for segment in conversation:
if 'voice' not in segment:
name = segment['name']
if name in name_to_voice:
voice = name_to_voice[name]
else:
if segment['gender'].lower() == 'male':
voice = random.choice(MALE_NEURAL_VOICES)['Id']
else:
voice = random.choice(FEMALE_NEURAL_VOICES)['Id']
name_to_voice[name] = voice
segment['voice'] = voice
# Initialize the Amazon Polly client
client = boto3.client(
'polly',
@@ -77,7 +64,6 @@ def conversation_text_to_speech(conversation: list, file_name: str):
audio_segments.append(response['AudioStream'].read())
# Combine the audio segments into a single audio file
combined_audio = b"".join(audio_segments)
file_name = file_name + ".mp3"
# Save the combined audio to a single file
with open(file_name, "wb") as f:
f.write(combined_audio)
@@ -94,3 +80,24 @@ def has_10_words(text: str):
words_in_input = text.split()
english_word_count = sum(1 for word in words_in_input if word.lower() in english_words)
return english_word_count >= 10
def divide_text(text, max_length=3000):
if len(text) <= max_length:
return [text]
divisions = []
current_position = 0
while current_position < len(text):
next_position = min(current_position + max_length, len(text))
next_period_position = text.rfind('.', current_position, next_position)
if next_period_position != -1 and next_period_position > current_position:
divisions.append(text[current_position:next_period_position + 1])
current_position = next_period_position + 1
else:
# If no '.' found in the next chunk, split at max_length
divisions.append(text[current_position:next_position])
current_position = next_position
return divisions

1595
templates/question_templates.py
View File

File diff suppressed because it is too large Load Diff