Pipecat Krisp
This module provides a Python wrapper around Krisp's C++ SDK, enabling its integration with Pipecat. Krisp offers a Desktop SDK (in C++) that supports Linux, Windows, and macOS, allowing applications to leverage noise suppression capabilities.
Prerequisites
- Krisp Desktop SDK: Download Krisp's Desktop SDK and the model you wish to use from Krisp's desktop SDKs.
- Environment Configuration:
- Copy the provided
env_templatefile to a new file named.env. - In
.env, set the paths to the downloaded Krisp SDK and model files.
- Copy the provided
- CPython Module Dependencies
- In addition to above dependencies you will need the pybind11 library to build CPython modules using C++.
- On Ubuntu Linux:
sudo apt-get install pybind11-dev - On Mac:
brew install pybind11
Environment Template
Below is the format for the .env file. Replace the placeholders with the actual paths where the Krisp SDK and model are located:
KRISP_SDK_PATH=/PATH/TO/KRISP/SDK
KRISP_MODEL_PATH=/PATH/TO/KRISP/MODEL
Installation and Setup
Follow these steps to set up and build the Pipecat Krisp Module:
- Set up a Virtual Environment:
python3 -m venv venv
source venv/bin/activate
- Install Dependencies:
pip install -r requirements.txt
- Build and Install the Local Python wrapper Library:
pip install --editable .
Running the Example
- Install Dependencies:
pip install -r example/requirements.txt
After setup, you can run an example script to test the Krisp integration:
./scripts/run_example.sh
An output folder will be created, containing the processed file with noise suppression applied by Krisp.
Publishing
- Prepare to dist:
./scripts/prepare_dist.sh
- Test the build using TestPyPI with Twine:
Upload to TestPyPI using twine
twine upload --repository testpypi dist/*
Test using pip to download packages from TestPyPI instead of PyPI
pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ pipecat-ai-krisp
Once you are happy, publish it to production.
twine upload dist/*
Use twine upload dist/* to upload your package and enter your credentials for the account you registered on the real PyPI. In this case you don’t need to specify --repository.
References
- Krisp getting started: https://sdk-docs.krisp.ai/docs/getting-started-js
- Krisp Sample Apps: https://github.com/krispai/Krisp-SDK-Sample-Apps/tree/krisp-sdk-v9
- Python packaging guide: https://packaging.python.org/en/latest/guides/using-testpypi/
Getting help
Metadata
Release files for pipecat-ai-krisp 0.4.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| pipecat_ai_krisp-0.4.0.tar.gz | 13.3 kB | Details |
Release files / pipecat_ai_krisp-0.4.0.tar.gz
| Download URL | pipecat_ai_krisp-0.4.0.tar.gz |
|---|---|
| Size | 13.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
4f0e05e218dcf15874957e9851299e219c713a0aa8353d2fd811f1b54001a602
|
|
BLAKE2b-256 checksum How to use checksums |
1d370f1d11d1dc33234a36de01992a9e5adc3c5e1dce71cc87b2bf909fa2f698
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.12.6
|