Simplify the interface to control model

Remove VoiceRequest and FaceRequest. Use List<Voice> and List<FaceExpression> instead.

Examples/ChatGPT/ChatGPTSkill.cs

@@ -37,11 +37,11 @@ private void Start()
             var processingAnimation = new List<Model.Animation>();
             processingAnimation.Add(new Model.Animation("BaseParam", 3, 0.3f));
             processingAnimation.Add(new Model.Animation("BaseParam", 3, 20.0f, "AGIA_Layer_nodding_once_01", "Additive Layer"));
-            var processingFace = new FaceRequest();
-            processingFace.AddFace("Blink", 3.0f);
+            var processingFace = new List<FaceExpression>();
+            processingFace.Add(new FaceExpression("Blink", 3.0f));
 
-            var neutralFaceRequest = new FaceRequest();
-            neutralFaceRequest.AddFace("Neutral");
+            var neutralFaceRequest = new List<FaceExpression>();
+            neutralFaceRequest.Add(new FaceExpression("Neutral"));
 
             var dialogController = gameObject.GetComponent<DialogController>();
             dialogController.OnRequestAsync = async (request, token) =>
@@ -60,8 +60,8 @@ private void Start()
             animationOnStart.Add(new Model.Animation("BaseParam", 10, 3.0f));
             modelController.Animate(animationOnStart);
 
-            var faceOnStart = new FaceRequest();
-            faceOnStart.AddFace("Joy", 3.0f);
+            var faceOnStart = new List<FaceExpression>();
+            faceOnStart.Add(new FaceExpression("Joy", 3.0f));
             modelController.SetFace(faceOnStart);
         }
 

Scripts/Model/FaceExpression.cs

@@ -7,11 +7,11 @@ public class FaceExpression
         public float Duration { get; set; }
         public string Description { get; set; }
 
-        public FaceExpression(string name, float duration, string description)
+        public FaceExpression(string name, float duration = 0.0f, string description = null)
         {
             Name = name;
             Duration = duration;
-            Description = description;
+            Description = description ?? string.Empty;
         }
     }
 }

Scripts/Model/ModelController.cs

@@ -16,7 +16,7 @@ public class ModelController : MonoBehaviour
         // Audio
         [Header("Voice")]
         public AudioSource AudioSource;
-        private Dictionary<string, AudioClip> voices = new Dictionary<string, AudioClip>();
+        private Dictionary<string, AudioClip> voiceAudioClips = new Dictionary<string, AudioClip>();
         public Func<Voice, CancellationToken, UniTask<AudioClip>> VoiceDownloadFunc;
         public Func<Voice, CancellationToken, UniTask<AudioClip>> TextToSpeechFunc;
         public Dictionary<string, Func<Voice, CancellationToken, UniTask<AudioClip>>> TextToSpeechFunctions = new Dictionary<string, Func<Voice, CancellationToken, UniTask<AudioClip>>>();
@@ -134,7 +134,7 @@ public void AddIdleAnimation(Animation animation, int weight = 1)
                 idleWeightedIndexes.Add(index);
             }
         }
-        #endregion
+#endregion
 
         // Speak with animation and face expression
         public async UniTask AnimatedSay(AnimatedVoiceRequest request, CancellationToken token)
@@ -156,14 +156,14 @@ public async UniTask AnimatedSay(AnimatedVoiceRequest request, CancellationToken
                 // Face
                 if (animatedVoice.Faces != null && animatedVoice.Faces.Count > 0)
                 {
-                    SetFace(new FaceRequest(animatedVoice.Faces));
+                    SetFace(animatedVoice.Faces);
                 }
 
                 // Speech
                 if (animatedVoice.Voices.Count > 0)
                 {
                     // Wait for the requested voices end
-                    await Say(new VoiceRequest(animatedVoice.Voices), token);
+                    await Say(animatedVoice.Voices, token);
                 }
             }
 
@@ -176,29 +176,20 @@ public async UniTask AnimatedSay(AnimatedVoiceRequest request, CancellationToken
         }
 
 #region Speech
-        // Speak one phrase
-        public async UniTask Say(string voiceName, float preGap = 0f, float postGap = 0f)
-        {
-            var request = new VoiceRequest(voiceName);
-            request.Voices[0].PreGap = preGap;
-            request.Voices[0].PostGap = postGap;
-            await Say(request, new CancellationTokenSource().Token); // ノンキャンセラブル
-        }
-
         // Speak
-        public async UniTask Say(VoiceRequest request, CancellationToken token)
+        public async UniTask Say(List<Voice> voices, CancellationToken token)
         {
             // Stop speech
             StopSpeech();
 
             // Prefetch Web/TTS voice
             if (UsePrefetch)
             {
-                PrefetchVoices(request.Voices, token);
+                PrefetchVoices(voices, token);
             }
 
             // Speak sequentially
-            foreach (var v in request.Voices)
+            foreach (var v in voices)
             {
                 if (token.IsCancellationRequested)
                 {
@@ -207,13 +198,13 @@ public async UniTask Say(VoiceRequest request, CancellationToken token)
 
                 if (v.Source == VoiceSource.Local)
                 {
-                    if (voices.ContainsKey(v.Name))
+                    if (voiceAudioClips.ContainsKey(v.Name))
                     {
                         // Wait for PreGap
                         await UniTask.Delay((int)(v.PreGap * 1000), cancellationToken: token);
                         // Play audio
                         History?.Add(v);
-                        AudioSource.PlayOneShot(voices[v.Name]);
+                        AudioSource.PlayOneShot(voiceAudioClips[v.Name]);
                     }
                     else
                     {
@@ -325,7 +316,7 @@ public void StopSpeech()
         // Register voice with its name
         public void AddVoice(string name, AudioClip audioClip)
         {
-            voices[ReplaceDakuten(name)] = audioClip;
+            voiceAudioClips[ReplaceDakuten(name)] = audioClip;
         }
 
         // Get registered TTS Function by name
@@ -376,7 +367,7 @@ private string ReplaceDakuten(string value)
 
             return ret;
         }
-        #endregion
+#endregion
 
 
 #region Animation
@@ -450,10 +441,10 @@ private Animation GetIdleAnimation()
 
 #region Face Expression
         // Set face expressions
-        public void SetFace(FaceRequest request)
+        public void SetFace(List<FaceExpression> faces)
         {
             faceQueue.Clear();
-            faceQueue = new List<FaceExpression>(request.Faces);
+            faceQueue = new List<FaceExpression>(faces);    // Copy faces not to change original list
             faceStartAt = 0;
         }
 
@@ -464,7 +455,7 @@ private void UpdateFace()
             if (faceToSet == null)
             {
                 // Set neutral instead when faceToSet is null
-                SetFace(new FaceRequest(new List<FaceExpression>() { new FaceExpression("Neutral", 0.0f, string.Empty) }));
+                SetFace(new List<FaceExpression>() { new FaceExpression("Neutral", 0.0f, string.Empty) });
                 return;
             }
 

Tests/TestModelRequests.cs

@@ -67,73 +67,6 @@ public void TestVoice()
             Assert.AreEqual("tts voice", ttsVoice.Description);
         }
 
-        [Test]
-        public void TestVoiceRequest()
-        {
-            // Without params
-            var voiceRequest = new VoiceRequest();
-
-            voiceRequest.AddVoice("LocalVoice");
-            voiceRequest.AddVoiceWeb("https://voice.url");
-            voiceRequest.AddVoiceTTS("This text will be read by chatdoll.");
-
-            var localVoice = voiceRequest.Voices[0];
-            Assert.AreEqual("LocalVoice", localVoice.Name);
-            Assert.AreEqual(0.0f, localVoice.PreGap);
-            Assert.AreEqual(0.0f, localVoice.PostGap);
-            Assert.AreEqual(string.Empty, localVoice.Text);
-            Assert.AreEqual(string.Empty, localVoice.Url);
-            Assert.IsNull(localVoice.TTSConfig);
-            Assert.AreEqual(VoiceSource.Local, localVoice.Source);
-
-            var webVoice = voiceRequest.Voices[1];
-            Assert.AreEqual(string.Empty, webVoice.Name);
-            Assert.AreEqual(0.0f, webVoice.PreGap);
-            Assert.AreEqual(0.0f, webVoice.PostGap);
-            Assert.AreEqual(string.Empty, webVoice.Text);
-            Assert.AreEqual("https://voice.url", webVoice.Url);
-            Assert.IsNull(webVoice.TTSConfig);
-            Assert.AreEqual(VoiceSource.Web, webVoice.Source);
-            Assert.IsTrue(webVoice.UseCache);
-
-            var ttsVoice = voiceRequest.Voices[2];
-            Assert.AreEqual(string.Empty, ttsVoice.Name);
-            Assert.AreEqual(0.0f, ttsVoice.PreGap);
-            Assert.AreEqual(0.0f, ttsVoice.PostGap);
-            Assert.AreEqual("This text will be read by chatdoll.", ttsVoice.Text);
-            Assert.AreEqual(string.Empty, ttsVoice.Url);
-            Assert.IsNull(ttsVoice.TTSConfig);
-            Assert.AreEqual(VoiceSource.TTS, ttsVoice.Source);
-            Assert.IsTrue(ttsVoice.UseCache);
-
-            // With params
-            var voiceRequestP = new VoiceRequest();
-
-            voiceRequestP.AddVoice("LocalVoice", 0.1f, 0.2f);
-            voiceRequestP.AddVoiceWeb("https://voice.url", 0.1f, 0.2f, useCache: false);
-            var ttsConfig = new TTSConfiguration("TestTTSFuncName");
-            ttsConfig.Params.Add("key1", "val1");
-            ttsConfig.Params.Add("key2", 2.0f);
-            voiceRequestP.AddVoiceTTS("This text will be read by chatdoll.", 0.1f, 0.2f, ttsConfig: ttsConfig, useCache: false);
-
-            var localVoiceP = voiceRequestP.Voices[0];
-            Assert.AreEqual(0.1f, localVoiceP.PreGap);
-            Assert.AreEqual(0.2f, localVoiceP.PostGap);
-
-            var webVoiceP = voiceRequestP.Voices[1];
-            Assert.AreEqual(0.1f, webVoiceP.PreGap);
-            Assert.AreEqual(0.2f, webVoiceP.PostGap);
-            Assert.IsFalse(webVoiceP.UseCache);
-
-            var ttsVoiceP = voiceRequestP.Voices[2];
-            Assert.AreEqual(0.1f, ttsVoiceP.PreGap);
-            Assert.AreEqual(0.2f, ttsVoiceP.PostGap);
-            Assert.AreEqual("TestTTSFuncName", ttsVoiceP.GetTTSFunctionName());
-            Assert.AreEqual("val1", ttsVoiceP.GetTTSParam("key1"));
-            Assert.AreEqual(2.0f, ttsVoiceP.GetTTSParam("key2"));
-            Assert.IsFalse(ttsVoiceP.UseCache);
-        }
-
         [Test]
         public void TestFaceClip()
         {
@@ -203,22 +136,6 @@ public void TestFaceExpression()
             Assert.AreEqual("face test", faceExpression.Description);
         }
 
-        [Test]
-        public void TestFaceRequest()
-        {
-            var faceRequest = new FaceRequest();
-            faceRequest.AddFace("Smile");
-            faceRequest.AddFace("Angry", 1.1f, "angry face");
-
-            Assert.AreEqual("Smile", faceRequest.Faces[0].Name);
-            Assert.AreEqual(0.0f, faceRequest.Faces[0].Duration);
-            Assert.AreEqual(null, faceRequest.Faces[0].Description);
-
-            Assert.AreEqual("Angry", faceRequest.Faces[1].Name);
-            Assert.AreEqual(1.1f, faceRequest.Faces[1].Duration);
-            Assert.AreEqual("angry face", faceRequest.Faces[1].Description);
-        }
-
         [Test]
         public void TestAnimation()
         {