回答編集履歴
2
内容追記
answer
CHANGED
|
@@ -19,4 +19,179 @@
|
|
|
19
19
|
Qiita | UnityでMicrosoft Cognitive Speech Servicesによる音声認識をAndroidスマホで実装する
|
|
20
20
|
[https://qiita.com/tomato_sugar/items/ac4b5dbe8277496add56](https://qiita.com/tomato_sugar/items/ac4b5dbe8277496add56)
|
|
21
21
|
|
|
22
|
-
途中でわからないことがあれば言ってください。アドバイスできると思います。
|
|
22
|
+
途中でわからないことがあれば言ってください。アドバイスできると思います。
|
|
23
|
+
|
|
24
|
+
---
|
|
25
|
+
|
|
26
|
+
追記：Azure Cognitive Speech SDK 導入方法
|
|
27
|
+
|
|
28
|
+
自分のアカウントのAzure上で Cognitive Services を立てている前提です。
|
|
29
|
+
立て方は上記のQiita記事を参考にしてください。
|
|
30
|
+
|
|
31
|
+
1, Microsoftさん公式のUnityのサンプルプロジェクトをgit cloneする。
|
|
32
|
+
[https://github.com/Azure-Samples/cognitive-services-speech-sdk/tree/master/quickstart/csharp/unity/from-microphone](https://github.com/Azure-Samples/cognitive-services-speech-sdk/tree/master/quickstart/csharp/unity/from-microphone)
|
|
33
|
+
※これをzipでDLして quickstart/csharp/unity/from-microphone でもよい。
|
|
34
|
+
[https://github.com/Azure-Samples/cognitive-services-speech-sdk](https://github.com/Azure-Samples/cognitive-services-speech-sdk)
|
|
35
|
+
|
|
36
|
+
2, `Microsoft.CognitiveServices.Speech.xxx.unitypackage` をDLしてインポートする[https://aka.ms/csspeech/unitypackage](https://aka.ms/csspeech/unitypackage)
|
|
37
|
+
|
|
38
|
+
3, Assets/Scripts/`HelloWorld.cs` を下の内容に書き換える。
|
|
39
|
+
変更点は、キー等の情報をスクリプトに直接記入するようになっていたので[SerializeField]でEditor上から入力できるようにした。あと日本語対応。
|
|
40
|
+
```C#
|
|
41
|
+
//
|
|
42
|
+
// Copyright (c) Microsoft. All rights reserved.
|
|
43
|
+
// Licensed under the MIT license. See LICENSE.md file in the project root for full license information.
|
|
44
|
+
//
|
|
45
|
+
// <code>
|
|
46
|
+
using UnityEngine;
|
|
47
|
+
using UnityEngine.UI;
|
|
48
|
+
using Microsoft.CognitiveServices.Speech;
|
|
49
|
+
#if PLATFORM_ANDROID
|
|
50
|
+
using UnityEngine.Android;
|
|
51
|
+
#endif
|
|
52
|
+
#if PLATFORM_IOS
|
|
53
|
+
using UnityEngine.iOS;
|
|
54
|
+
using System.Collections;
|
|
55
|
+
#endif
|
|
56
|
+
|
|
57
|
+
public class HelloWorld : MonoBehaviour
|
|
58
|
+
{
|
|
59
|
+
// Hook up the two properties below with a Text and Button object in your UI.
|
|
60
|
+
public Text outputText;
|
|
61
|
+
public Button startRecoButton;
|
|
62
|
+
|
|
63
|
+
private object threadLocker = new object();
|
|
64
|
+
private bool waitingForReco;
|
|
65
|
+
private string message;
|
|
66
|
+
|
|
67
|
+
private bool micPermissionGranted = false;
|
|
68
|
+
|
|
69
|
+
#if PLATFORM_ANDROID || PLATFORM_IOS
|
|
70
|
+
// Required to manifest microphone permission, cf.
|
|
71
|
+
// https://docs.unity3d.com/Manual/android-manifest.html
|
|
72
|
+
private Microphone mic;
|
|
73
|
+
#endif
|
|
74
|
+
|
|
75
|
+
[SerializeField] string subscriptionKey = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx";
|
|
76
|
+
[SerializeField] string serviceRegion = "japaneast";
|
|
77
|
+
|
|
78
|
+
public async void ButtonClick()
|
|
79
|
+
{
|
|
80
|
+
// Creates an instance of a speech config with specified subscription key and service region.
|
|
81
|
+
// Replace with your own subscription key and service region (e.g., "westus").
|
|
82
|
+
var config = SpeechConfig.FromSubscription(subscriptionKey, serviceRegion);
|
|
83
|
+
var lang = SourceLanguageConfig.FromLanguage("ja-JP");
|
|
84
|
+
|
|
85
|
+
// Make sure to dispose the recognizer after use!
|
|
86
|
+
using (var recognizer = new SpeechRecognizer(config, lang))
|
|
87
|
+
{
|
|
88
|
+
lock (threadLocker)
|
|
89
|
+
{
|
|
90
|
+
waitingForReco = true;
|
|
91
|
+
}
|
|
92
|
+
|
|
93
|
+
// Starts speech recognition, and returns after a single utterance is recognized. The end of a
|
|
94
|
+
// single utterance is determined by listening for silence at the end or until a maximum of 15
|
|
95
|
+
// seconds of audio is processed. The task returns the recognition text as result.
|
|
96
|
+
// Note: Since RecognizeOnceAsync() returns only a single utterance, it is suitable only for single
|
|
97
|
+
// shot recognition like command or query.
|
|
98
|
+
// For long-running multi-utterance recognition, use StartContinuousRecognitionAsync() instead.
|
|
99
|
+
var result = await recognizer.RecognizeOnceAsync().ConfigureAwait(false);
|
|
100
|
+
|
|
101
|
+
// Checks result.
|
|
102
|
+
string newMessage = string.Empty;
|
|
103
|
+
if (result.Reason == ResultReason.RecognizedSpeech)
|
|
104
|
+
{
|
|
105
|
+
newMessage = result.Text;
|
|
106
|
+
}
|
|
107
|
+
else if (result.Reason == ResultReason.NoMatch)
|
|
108
|
+
{
|
|
109
|
+
newMessage = "NOMATCH: Speech could not be recognized.";
|
|
110
|
+
}
|
|
111
|
+
else if (result.Reason == ResultReason.Canceled)
|
|
112
|
+
{
|
|
113
|
+
var cancellation = CancellationDetails.FromResult(result);
|
|
114
|
+
newMessage = $"CANCELED: Reason={cancellation.Reason} ErrorDetails={cancellation.ErrorDetails}";
|
|
115
|
+
}
|
|
116
|
+
|
|
117
|
+
lock (threadLocker)
|
|
118
|
+
{
|
|
119
|
+
message = newMessage;
|
|
120
|
+
waitingForReco = false;
|
|
121
|
+
}
|
|
122
|
+
}
|
|
123
|
+
}
|
|
124
|
+
|
|
125
|
+
void Start()
|
|
126
|
+
{
|
|
127
|
+
if (outputText == null)
|
|
128
|
+
{
|
|
129
|
+
UnityEngine.Debug.LogError("outputText property is null! Assign a UI Text element to it.");
|
|
130
|
+
}
|
|
131
|
+
else if (startRecoButton == null)
|
|
132
|
+
{
|
|
133
|
+
message = "startRecoButton property is null! Assign a UI Button to it.";
|
|
134
|
+
UnityEngine.Debug.LogError(message);
|
|
135
|
+
}
|
|
136
|
+
else
|
|
137
|
+
{
|
|
138
|
+
// Continue with normal initialization, Text and Button objects are present.
|
|
139
|
+
#if PLATFORM_ANDROID
|
|
140
|
+
// Request to use the microphone, cf.
|
|
141
|
+
// https://docs.unity3d.com/Manual/android-RequestingPermissions.html
|
|
142
|
+
message = "Waiting for mic permission";
|
|
143
|
+
if (!Permission.HasUserAuthorizedPermission(Permission.Microphone))
|
|
144
|
+
{
|
|
145
|
+
Permission.RequestUserPermission(Permission.Microphone);
|
|
146
|
+
}
|
|
147
|
+
#elif PLATFORM_IOS
|
|
148
|
+
if (!Application.HasUserAuthorization(UserAuthorization.Microphone))
|
|
149
|
+
{
|
|
150
|
+
Application.RequestUserAuthorization(UserAuthorization.Microphone);
|
|
151
|
+
}
|
|
152
|
+
#else
|
|
153
|
+
micPermissionGranted = true;
|
|
154
|
+
message = "Click button to recognize speech";
|
|
155
|
+
#endif
|
|
156
|
+
startRecoButton.onClick.AddListener(ButtonClick);
|
|
157
|
+
}
|
|
158
|
+
}
|
|
159
|
+
|
|
160
|
+
void Update()
|
|
161
|
+
{
|
|
162
|
+
#if PLATFORM_ANDROID
|
|
163
|
+
if (!micPermissionGranted && Permission.HasUserAuthorizedPermission(Permission.Microphone))
|
|
164
|
+
{
|
|
165
|
+
micPermissionGranted = true;
|
|
166
|
+
message = "Click button to recognize speech";
|
|
167
|
+
}
|
|
168
|
+
#elif PLATFORM_IOS
|
|
169
|
+
if (!micPermissionGranted && Application.HasUserAuthorization(UserAuthorization.Microphone))
|
|
170
|
+
{
|
|
171
|
+
micPermissionGranted = true;
|
|
172
|
+
message = "Click button to recognize speech";
|
|
173
|
+
}
|
|
174
|
+
#endif
|
|
175
|
+
|
|
176
|
+
lock (threadLocker)
|
|
177
|
+
{
|
|
178
|
+
if (startRecoButton != null)
|
|
179
|
+
{
|
|
180
|
+
startRecoButton.interactable = !waitingForReco && micPermissionGranted;
|
|
181
|
+
}
|
|
182
|
+
if (outputText != null)
|
|
183
|
+
{
|
|
184
|
+
outputText.text = message;
|
|
185
|
+
}
|
|
186
|
+
}
|
|
187
|
+
}
|
|
188
|
+
}
|
|
189
|
+
// </code>
|
|
190
|
+
|
|
191
|
+
```
|
|
192
|
+
|
|
193
|
+
4, 作成済みのリソースから `キー1` を `subscriptionKey` に張り付ける。また `場所` が `serviceRegion` に相当するので japaneast でなければ書き換える。
|
|
194
|
+
|
|
195
|
+
|
|
196
|
+
5, Build Settings の Target Platform を Android に変更してビルド。実機で動作確認。（Windows でも iOS でもOKです。）
|
|
197
|
+
|
1
書式修正
answer
CHANGED
|
@@ -1,9 +1,9 @@
|
|
|
1
|
-
>音声認識APIを用いて音声をテキスト化し、switch文の中で照合させて合致したcase内の処理によってアニメーションを変えるという流れなのでしょうか
|
|
1
|
+
> 音声認識APIを用いて音声をテキスト化し、switch文の中で照合させて合致したcase内の処理によってアニメーションを変えるという流れなのでしょうか
|
|
2
2
|
|
|
3
3
|
その処理で大丈夫です。
|
|
4
4
|
|
|
5
5
|
|
|
6
|
-
>『そもそもAPIをどの様に用いれば上記の処理が出来るのか』が分からない
|
|
6
|
+
> 『そもそもAPIをどの様に用いれば上記の処理が出来るのか』が分からない
|
|
7
7
|
|
|
8
8
|
unityで利用できる音声認識APIは以下のようなものがあります。
|
|
9
9
|
・`UnityEngine.Windows.Speech`
|