Metadata-Version: 2.4
Name: ovos-vad-plugin-noise
Version: 0.1.3a5
Summary: noise ratio VAD plugin for OpenVoiceOS
Author-email: JarbasAi <jarbasai@mailfence.com>
License: Apache-2.0
Project-URL: Homepage, https://github.com/OpenVoiceOS/ovos-vad-plugin-noise
Project-URL: Repository, https://github.com/OpenVoiceOS/ovos-vad-plugin-noise
Keywords: vad,noise,ovos,openvoiceos,plugin,voice-activity-detection
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: ovos-plugin-manager>=2.4.0a1
Provides-Extra: test
Requires-Dist: pytest<9; extra == "test"
Requires-Dist: ovoscope<1.0.0,>=0.19.0a1; extra == "test"
Requires-Dist: ovos-dinkum-listener; extra == "test"
Dynamic: license-file

# Noise VAD

simple VAD plugin extracted from the old [ovos-listener](https://github.com/OpenVoiceOS/ovos-listener/blob/dev/ovos_listener/silence.py)

should only be used as fallback, works in all platforms including 32bit systems

## Configuration


```javascript
{
    "listener": {
        "VAD": {
            "module": "ovos-vad-plugin-noise",
            "ovos-vad-plugin-noise": {
                "method": "all",
                "max_current_ratio_threshold": 2.0,
                "energy_threshold": 1000.0
            }
        }
    }
}
```

Arguments

    max_energy: Optional[float] = None
        Maximum denoise energy value (None for dynamic setting from observed audio)

    max_current_ratio_threshold: Optional[float] = 2.0
        Ratio of max/current energy below which audio is considered speech

    energy_threshold: Optional[float] = None
        Energy threshold above which audio is considered speech (None for dynamic setting from observed audio)

    silence_method: SilenceMethod = "all"
        Method for deciding if an audio chunk contains silence or speech

Methods

    RATIO
      Only use max/current energy ratio threshold

    THRESHOLD
      Only use current energy threshold

    ALL
      max/current energy ratio, and current energy threshold
