# SpeakerVerSim [](https://github.com/wq2012/SpeakerVerSim/actions/workflows/python-app.yml) [](https://pypi.python.org/pypi/SpeakerVerSim) [](https://pypi.org/project/SpeakerVerSim) [](https://pepy.tech/project/SpeakerVerSim)
## Overview
SpeakerVerSim is an easily-extensible Python-based simulation framework for different version control strategies of speaker recognition systems under different network configurations.
These simulations are used in the paper [Version Control of Speaker Recognition Systems](https://www.sciencedirect.com/science/article/pii/S0164121224001675) ([arXiv version](https://arxiv.org/abs/2007.12069)).
## How to use
### Install
The `SpeakerVerSim` library can be installed with:
```
pip install SpeakerVerSim
```
### Run one simulation
You can easily start a simulation by running:
```
python run_simulator.py
```
This simulation will use the configurations in the `example_config.yml` file.
To use a different configuration file, you can use the `-c` or `--config` flag. For example:
```
python run_simulator.py --config my_config.yml
```
You can also override the strategy to be simulated from the config file using the `-s` or `--strategy` flag. For example:
```
python run_simulator.py -c my_config.yml -s SSO-sync
```
### Call the API
The highest level API is the `SpeakerVerSim.simulate` function.
It takes either the path to the config file or an object of the configurations as its input, and outputs the simulation stats.
Example usage:
```
import SpeakerVerSim
stats = SpeakerVerSim.simulate("example_config.yml")
```
### Reproduce experiments
You can easily reproduce the experiments in our [paper](https://arxiv.org/abs/2007.12069) by running:
```
python run_exp.py
```
After running this script, simulation results will be stored in the `result_stats` directory.
Then you can visualize the metrics by running:
```
python visualize_results.py
```
The visualization graphics will be stored in the `figures` directory.
## List of implemented strategies
| Script | Strategy | Description |
| ------------------------------- | ----------- | ----------- |
| `server_single_simple.py` | SSO | Basic server-side single version online updating strategy.
| `server_single_sync.py` | SSO-sync | Server-side single version online updating strategy with frontend-worker version sync.
| `server_single_hash.py` | SSO-hash | Server-side single version online updating strategy with user-ID hashing.
| `server_single_multiprofile.py` | SSO-mul | Server-side single version online updating strategy with multi-profile database.
| `server_single_sync.py` | SD | Server-side double version updating strategy.
## Design
The design of this library is summarized as below:
* This library is built on top of [SimPy](https://simpy.readthedocs.io), a process-based discrete-event simulation (DES) framework based on standard Python.
* All configurations of the simulation are represented in a single YAML file. `example_config.yml` has explanations for all the configuration fields.
* Each machine in the network inherits from the `Actor` class, including the client, the frontend server, the cloud worker, and the database.
* All clients inherit from the `BaseClient` class; all frontend servers inherit from the `BaseFrontend` class; all cloud workers inherit from the `BaseWorker` class; and all databases inherit from the `BaseDatabase` class.
* The communication between two machines happens like this: the sender creates a `Message` object, and adds it to the receiver's message pool, which is a `simpy.Store` object.
* During the simulation, metrics are logged in an object of the `GlobalStats` class.
* The entire network system is represented by the `NetworkSystem` class or its subclass.
Each version control strategy is implemented by creating a set of client, frontend server, cloud workers, database, and defining how they interact with each other.
## Citation
```
@article{WANG2024112122,
title = {Version control of speaker recognition systems},
journal = {Journal of Systems and Software},
volume = {216},
pages = {112122},
year = {2024},
issn = {0164-1212},
doi = {https://doi.org/10.1016/j.jss.2024.112122},
url = {https://www.sciencedirect.com/science/article/pii/S0164121224001675},
author = {Quan Wang and Ignacio {Lopez Moreno}},
}
```
Raw data
{
"_id": null,
"home_page": "https://github.com/wq2012/SpeakerVerSim",
"name": "SpeakerVerSim",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": null,
"author": "Quan Wang",
"author_email": "quanw@google.com",
"download_url": "https://files.pythonhosted.org/packages/76/ef/90bb735b854901bde98f00825bd0c59438a1a17ee8c7ae07fbd66ad1d40c/speakerversim-0.1.3.tar.gz",
"platform": null,
"description": "# SpeakerVerSim [](https://github.com/wq2012/SpeakerVerSim/actions/workflows/python-app.yml) [](https://pypi.python.org/pypi/SpeakerVerSim) [](https://pypi.org/project/SpeakerVerSim) [](https://pepy.tech/project/SpeakerVerSim)\n\n\n\n## Overview\nSpeakerVerSim is an easily-extensible Python-based simulation framework for different version control strategies of speaker recognition systems under different network configurations.\n\nThese simulations are used in the paper [Version Control of Speaker Recognition Systems](https://www.sciencedirect.com/science/article/pii/S0164121224001675) ([arXiv version](https://arxiv.org/abs/2007.12069)).\n\n## How to use\n\n### Install\n\nThe `SpeakerVerSim` library can be installed with:\n\n```\npip install SpeakerVerSim\n```\n\n### Run one simulation\n\nYou can easily start a simulation by running:\n\n```\npython run_simulator.py\n```\n\nThis simulation will use the configurations in the `example_config.yml` file.\n\nTo use a different configuration file, you can use the `-c` or `--config` flag. For example:\n\n```\npython run_simulator.py --config my_config.yml\n```\n\nYou can also override the strategy to be simulated from the config file using the `-s` or `--strategy` flag. For example:\n\n```\npython run_simulator.py -c my_config.yml -s SSO-sync\n```\n\n### Call the API\n\nThe highest level API is the `SpeakerVerSim.simulate` function.\nIt takes either the path to the config file or an object of the configurations as its input, and outputs the simulation stats.\n\nExample usage:\n\n```\nimport SpeakerVerSim\n\nstats = SpeakerVerSim.simulate(\"example_config.yml\")\n```\n\n### Reproduce experiments\n\nYou can easily reproduce the experiments in our [paper](https://arxiv.org/abs/2007.12069) by running:\n\n```\npython run_exp.py\n```\n\nAfter running this script, simulation results will be stored in the `result_stats` directory.\n\nThen you can visualize the metrics by running:\n\n```\npython visualize_results.py\n```\n\nThe visualization graphics will be stored in the `figures` directory.\n\n## List of implemented strategies\n\n| Script | Strategy | Description |\n| ------------------------------- | ----------- | ----------- |\n| `server_single_simple.py` | SSO | Basic server-side single version online updating strategy.\n| `server_single_sync.py` | SSO-sync | Server-side single version online updating strategy with frontend-worker version sync.\n| `server_single_hash.py` | SSO-hash | Server-side single version online updating strategy with user-ID hashing.\n| `server_single_multiprofile.py` | SSO-mul | Server-side single version online updating strategy with multi-profile database.\n| `server_single_sync.py` | SD | Server-side double version updating strategy.\n\n## Design\n\nThe design of this library is summarized as below:\n\n* This library is built on top of [SimPy](https://simpy.readthedocs.io), a process-based discrete-event simulation (DES) framework based on standard Python.\n* All configurations of the simulation are represented in a single YAML file. `example_config.yml` has explanations for all the configuration fields.\n* Each machine in the network inherits from the `Actor` class, including the client, the frontend server, the cloud worker, and the database.\n* All clients inherit from the `BaseClient` class; all frontend servers inherit from the `BaseFrontend` class; all cloud workers inherit from the `BaseWorker` class; and all databases inherit from the `BaseDatabase` class.\n* The communication between two machines happens like this: the sender creates a `Message` object, and adds it to the receiver's message pool, which is a `simpy.Store` object.\n* During the simulation, metrics are logged in an object of the `GlobalStats` class.\n* The entire network system is represented by the `NetworkSystem` class or its subclass.\n\nEach version control strategy is implemented by creating a set of client, frontend server, cloud workers, database, and defining how they interact with each other.\n\n## Citation\n\n```\n@article{WANG2024112122,\n title = {Version control of speaker recognition systems},\n journal = {Journal of Systems and Software},\n volume = {216},\n pages = {112122},\n year = {2024},\n issn = {0164-1212},\n doi = {https://doi.org/10.1016/j.jss.2024.112122},\n url = {https://www.sciencedirect.com/science/article/pii/S0164121224001675},\n author = {Quan Wang and Ignacio {Lopez Moreno}},\n}\n```\n",
"bugtrack_url": null,
"license": null,
"summary": "Simulation framework for version control strategies of speaker recognition systems.",
"version": "0.1.3",
"project_urls": {
"Homepage": "https://github.com/wq2012/SpeakerVerSim"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8276cef3d46ff1b6ec88f432e340d149211ba67a50f9e5d082b859b7152f8aa2",
"md5": "d7d953c78e97cc969a52fe4283e93bb0",
"sha256": "1a7d1ce5b803b44cbc820abb9c6f978339eef9af42c2dd9b788164c2cd97b044"
},
"downloads": -1,
"filename": "SpeakerVerSim-0.1.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "d7d953c78e97cc969a52fe4283e93bb0",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 20675,
"upload_time": "2024-09-25T15:36:42",
"upload_time_iso_8601": "2024-09-25T15:36:42.400609Z",
"url": "https://files.pythonhosted.org/packages/82/76/cef3d46ff1b6ec88f432e340d149211ba67a50f9e5d082b859b7152f8aa2/SpeakerVerSim-0.1.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "76ef90bb735b854901bde98f00825bd0c59438a1a17ee8c7ae07fbd66ad1d40c",
"md5": "dbc267edd60df997694a926570d07711",
"sha256": "0e338629ca278ff5166313dfc14c0b6e8ab5615c4602c07fcb1bf383c8e0c445"
},
"downloads": -1,
"filename": "speakerversim-0.1.3.tar.gz",
"has_sig": false,
"md5_digest": "dbc267edd60df997694a926570d07711",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 16976,
"upload_time": "2024-09-25T15:36:43",
"upload_time_iso_8601": "2024-09-25T15:36:43.296187Z",
"url": "https://files.pythonhosted.org/packages/76/ef/90bb735b854901bde98f00825bd0c59438a1a17ee8c7ae07fbd66ad1d40c/speakerversim-0.1.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-09-25 15:36:43",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "wq2012",
"github_project": "SpeakerVerSim",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "simpy",
"specs": []
},
{
"name": "pyyaml",
"specs": []
},
{
"name": "tqdm",
"specs": []
},
{
"name": "seaborn",
"specs": []
},
{
"name": "munch",
"specs": []
}
],
"lcname": "speakerversim"
}
JSON
