Update documents

mochi-neko · Sep 5, 2023 · 93575fc · 93575fc
1 parent bc36b11
commit 93575fc
Show file tree

Hide file tree

Showing 6 changed files with 20 additions and 5 deletions.
diff --git a/Assets/Mochineko/VoiceActivityDetection.Samples/SampleVADParameters.asset b/Assets/Mochineko/VoiceActivityDetection.Samples/SampleVADParameters.asset
@@ -12,7 +12,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 2dffde2443bc4aa2a6d1155b1d78cd44, type: 3}
   m_Name: SampleVADParameters
   m_EditorClassIdentifier: 
-  activeVolumeThreshold: 0.005
+  activeVolumeThreshold: 0.01
   maxQueueingTimeSeconds: 0.5
   minQueueingTimeSeconds: 0.2
   activationRateThreshold: 0.6

diff --git a/Assets/Mochineko/VoiceActivityDetection.Samples/VADCamulationSample.cs b/Assets/Mochineko/VoiceActivityDetection.Samples/VADCamulationSample.cs
@@ -6,6 +6,9 @@
 
 namespace Mochineko.VoiceActivityDetection.Samples
 {
+    /// <summary>
+    /// A sample of voice activity detection by CumulativeVoiceActivityDetector.
+    /// </summary>
     internal sealed class VADCamulationSample : MonoBehaviour
     {
         [SerializeField]

diff --git a/Assets/Mochineko/VoiceActivityDetection/QueueingVoiceActivityDetector.cs b/Assets/Mochineko/VoiceActivityDetection/QueueingVoiceActivityDetector.cs
@@ -8,8 +8,7 @@
 namespace Mochineko.VoiceActivityDetection
 {
     /// <summary>
-    /// A simple implementation of <see cref="IVoiceActivityDetector"/>.
-    /// Detects voice activity by using voice segment queue, volume threshold, activation/deactivation rate and interval.
+    /// A voice activity detector that queues voice segments and detects activity by rate.
     /// </summary>
     public sealed class QueueingVoiceActivityDetector : IVoiceActivityDetector
     {

diff --git a/Assets/Mochineko/VoiceActivityDetection/package.json b/Assets/Mochineko/VoiceActivityDetection/package.json
@@ -1,6 +1,6 @@
 {
   "name": "com.mochineko.voice-activity-detection",
-  "version": "0.3.2",
+  "version": "0.4.0",
   "displayName": "Voice Activity Detection",
   "description": "A voice activity detection library for Unity.",
   "unity": "2022.3",

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.4.0] - 2023-09-05
+
+### Added
+- Add another VAD logic: `CumulativeVoiceActivityDetector`.
+- Add a sample of `CumulativeVoiceActivityDetector`.
+
+### Changed
+- Make `VoiceSegment` class to prevent struct copy and memory leak.
+
 ## [0.3.2] - 2023-09-05
 
 ### Fixed

diff --git a/README.md b/README.md
@@ -20,6 +20,9 @@ You can customize voice sources, voice buffers,
   - [x] AudioClip -> [AudioClipBuffer](./Assets/Mochineko/VoiceActivityDetection/AudioClipBuffer.cs)
 - Voice activity detection logics
   - [x] Queueing-based simple VAD logic -> [QueueingVoiceActivityDetector](./Assets/Mochineko/VoiceActivityDetection/QueueingVoiceActivityDetector.cs)
+    - Less memory usage but less stability 
+  - [x] Cumulative VAD logic -> [CumulativeVoiceActivityDetector](./Assets/Mochineko/VoiceActivityDetection/CumulativeVoiceActivityDetector.cs)
+    - More stability but more memory usage and less noise robustness
 
 ## How to import by UnityPackageManager
 
@@ -28,7 +31,7 @@ Add following dependencies to your `/Packages/manifest.json`.
 ```json
 {
     "dependencies": {
-        "com.mochineko.voice-activity-detection": "https://github.com/mochi-neko/voice-activity-detection-unity.git?path=/Assets/Mochineko/VoiceActivityDetection#0.3.2",
+        "com.mochineko.voice-activity-detection": "https://github.com/mochi-neko/voice-activity-detection-unity.git?path=/Assets/Mochineko/VoiceActivityDetection#0.4.0",
         "com.cysharp.unitask": "https://github.com/Cysharp/UniTask.git?path=src/UniTask/Assets/Plugins/UniTask",
         "com.neuecc.unirx": "https://github.com/neuecc/UniRx.git?path=Assets/Plugins/UniRx/Scripts",
         "com.naudio.core": "https://github.com/mochi-neko/simple-audio-codec-unity.git?path=/Assets/NAudio/NAudio.Core#0.2.0",
@@ -43,6 +46,7 @@ Add following dependencies to your `/Packages/manifest.json`.
 - [VAD with echo](./Assets/Mochineko/VoiceActivityDetection.Samples/VADAudioClipEchoSample.cs)
 - [VAD by AudioSource](./Assets/Mochineko/VoiceActivityDetection.Samples/VADAudioSourceSample.cs)
 - [VAD with OpenAI/Whisper API transcription](./Assets/Mochineko/VoiceActivityDetection.Samples/VADToWhisperSample.cs)
+- [VAD by cumulative logic](./Assets/Mochineko/VoiceActivityDetection.Samples/VADCamulationSample.cs)
 
 See also [Samples](./Assets/Mochineko/VoiceActivityDetection.Samples).