Call Center Transcription

​

Install Dependencies

pip install pyaudio 
pip install reverie-sdk
pip install python-dotenv

​

How does it Works?

1. Environment Setup:
   All sensitive credentials, such as API keys and application IDs, are securely loaded from environment variables. This ensures that sensitive information is not hardcoded into the script, promoting security and flexibility.

import os
import dotenv

dotenv.load_dotenv("../.env")
REVERIE_APP_ID = os.environ.get("REVERIE_APP_ID")
REVERIE_API_KEY = os.environ.get("REVERIE_API_KEY")

2. Client Initialization:
   A ReverieClient is initialized with the necessary credentials to interact with Reverie’s ASR services. The client is the primary interface through which the transcription and other API services are accessed.

from reverie_sdk import ReverieClient

client = ReverieClient(
    api_key=REVERIE_API_KEY,
    app_id=REVERIE_APP_ID,
)

3. Audio Processing and Transcription:
   The project is designed to handle various audio formats, processing them through Reverie’s ASR system. This includes batch processing for handling multiple files, real-time transcription, or other speech-to-text functionalities.

client.asr.stt_batch(
    "en",
    open("/home/rohnak.agarwal/Documents/sdk_sample/sample.en.rev.wav", "rb").read(),
)

4. Result Management: After transcribing the audio, the system outputs the transcription results. These results can then be stored, analyzed, or used for further processing depending on the use case (e.g., customer service analysis, data extraction, etc.).

for res in client.asr.stt_batch(
    "en",
    open("/home/rohnak.agarwal/Documents/sdk_sample/sample.en.rev.wav", "rb").read(),
):
    print(res)

​

Sample Code

import os
from reverie_sdk import ReverieClient
import dotenv

dotenv.load_dotenv("../.env")

REVERIE_APP_ID = os.environ.get("REVERIE_APP_ID")
REVERIE_API_KEY = os.environ.get("REVERIE_API_KEY")

client = ReverieClient(
    api_key=REVERIE_API_KEY,
    app_id=REVERIE_APP_ID,
)

for res in client.asr.stt_batch(
    "en",
    open("/home/rohnak.agarwal/Documents/sdk_sample/sample.en.rev.wav", "rb").read(),
):
    print(res)