This is the utils src code sub page!
This function is responsible for calling the agent, it is rate limited to 5 calls per minute, to avoid overloading the agent, and to avoid hitting rate limits on the provider side.
import streamlit as st
from ratelimit import limits
MINUTE = 60
@limits(calls=5, period=MINUTE)
def call(transcript):
"""The function `call` limits the number of calls to 5 per minute and processes a transcript to
generate a response.
Parameters
----------
transcript
The `transcript` parameter in the `call` function is expected to be a string representing the
user's input or conversation transcript that will be passed to the agent for processing.
Returns
-------
the text content of the last message in the response messages.
"""
response = st.session_state["agent"].invoke(
{"messages": [{"role": "user", "content": transcript}]}
)
return response["messages"][-1].content[-1]["text"]
These functions are responsible for caching the ElevenLabs client and the AI agent, to avoid re-initializing them every time they are called, which can be expensive in terms of time and resources.
from src.agent.agent import agent_pipeline
from elevenlabs.client import ElevenLabs
import streamlit as st
@st.cache_resource
def get_elevenlabs_client():
"""The function `get_elevenlabs_client` returns an instance of the ElevenLabs client using the API key
stored in the secrets.
Returns
-------
An instance of the ElevenLabs client with the API key retrieved from the secrets file.
"""
return ElevenLabs(api_key=st.secrets["ELEVENLABS_API_KEY"])
@st.cache_resource
def get_ai_agent():
"""The function `get_ai_agent` returns an AI agent created using a specified provider and model.
Returns
-------
The `get_ai_agent` function is returning an AI agent created using the `agent_pipeline` function
with the provider and model specified in the secrets file.
"""
return agent_pipeline(provider=st.secrets["PROVIDER"], model=st.secrets["MODEL"])
This function is responsible for converting speech to text using the ElevenLabs client, it takes an audio file as input and returns the transcribed text.
import streamlit as st
def speech_to_text(audio):
"""The function `speech_to_text` takes an audio file as input and uses a specified model to transcribe
the speech in the file, with options to tag audio events, specify language, and annotate speakers.
Parameters
----------
audio
The `audio` parameter in the `speech_to_text` function is the input audio file that you want to
convert to text. This file contains the speech that you want to transcribe into text.
Returns
-------
The function `speech_to_text` returns the transcription of the audio file provided as input, with
additional annotations such as tagging audio events like laughter, applause, etc., and diarization
to annotate who is speaking in the audio.
"""
transcription = st.session_state[
"elevenlabs"
].speech_to_text.convert(
file=audio,
model_id="scribe_v2", # Model to use
tag_audio_events=True, # Tag audio events like laughter, applause, etc.
language_code="eng", # Language of the audio file. If set to None, the model will detect the language automatically.
diarize=True, # Whether to annotate who is speaking
)
return transcription.text
This file contains utility functions for text processing, such as extracting text from PDF files and optimizing text by removing newlines and extra whitespace.
from pypdf import PdfReader
def extract_text(file) -> str:
"""The function `extract_text` reads text from a PDF file using the PdfReader library and returns the
concatenated text from all pages.
Parameters
----------
file
The `extract_text` function takes a file path as input and reads the text content from a PDF file
located at that path. It uses the `PdfReader` class to read the PDF file and extract text from each
page. The function then concatenates all the extracted text from each page into a
Returns
-------
The function `extract_text` returns a string containing the extracted text from all pages of the
PDF file specified as the input parameter.
"""
reader = PdfReader(file)
full_text = ""
for page in reader.pages:
full_text += page.extract_text()
return full_text
def optimizer(text: str) -> str:
"""The `optimizer` function in Python takes a string as input, removes any newline characters, and
returns the optimized string with leading and trailing whitespaces removed.
Parameters
----------
text : str
The `optimizer` function takes a string `text` as input, replaces any newline characters with
spaces, removes any leading or trailing whitespace, and then returns the optimized text.
Returns
-------
The `optimizer` function takes a string as input, replaces any newline characters with spaces,
removes any leading or trailing whitespace, and then returns the modified string.
"""
text = text.replace("\n", " ")
text = text.strip()
return text
This function is responsible for extracting text from PDF files, it takes a file path as input and returns the concatenated text from all pages.
def extract_text(file) -> str:
"""The function `extract_text` reads text from a PDF file using the PdfReader library and returns the
concatenated text from all pages.
Parameters
----------
file
The `extract_text` function takes a file path as input and reads the text content from a PDF file
located at that path. It uses the `PdfReader` class to read the PDF file and extract text from each
page. The function then concatenates all the extracted text from each page into a
Returns
-------
The function `extract_text` returns a string containing the extracted text from all pages of the
PDF file specified as the input parameter.
"""
reader = PdfReader(file)
full_text = ""
for page in reader.pages:
full_text += page.extract_text()
return full_text
This function is responsible for optimizing text by removing newlines and extra whitespace, it takes a string as input and returns the optimized string.
def optimizer(text: str) -> str:
"""The `optimizer` function in Python takes a string as input, removes any newline characters, and
returns the optimized string with leading and trailing whitespaces removed.
Parameters
----------
text : str
The `optimizer` function takes a string `text` as input, replaces any newline characters with
spaces, removes any leading or trailing whitespace, and then returns the optimized text.
Returns
-------
The `optimizer` function takes a string as input, replaces any newline characters with spaces,
removes any leading or trailing whitespace, and then returns the modified string.
"""
text = text.replace("\n", " ")
text = text.strip()
return text