添加回复用音频（文本转语音生成的）

语音会话添加play_reply方法实现播放回复
探索录音方式

Assets/training_speech/sd.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://dvvpoqb8umdjy"
+path="res://.godot/imported/sd.mp3-c6dc8a540e4c3c58e60586436e316423.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/sd.mp3"
+dest_files=["res://.godot/imported/sd.mp3-c6dc8a540e4c3c58e60586436e316423.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Assets/training_speech/xt_pslgbztmwx.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://iqr672ekex01"
+path="res://.godot/imported/xt_pslgbztmwx.mp3-c340c6f2fdc9b84c839f54346780b458.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/xt_pslgbztmwx.mp3"
+dest_files=["res://.godot/imported/xt_pslgbztmwx.mp3-c340c6f2fdc9b84c839f54346780b458.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Assets/training_speech/xt_pslibpkmwx.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://op4mbi1lemrb"
+path="res://.godot/imported/xt_pslibpkmwx.mp3-165b8ea64f29779308cdddb6c9efb934.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/xt_pslibpkmwx.mp3"
+dest_files=["res://.godot/imported/xt_pslibpkmwx.mp3-165b8ea64f29779308cdddb6c9efb934.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Assets/training_speech/zby_ztmjczc.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://dr1m5h02s45jf"
+path="res://.godot/imported/zby_ztmjczc.mp3-849bbce90b59ebabfadd754a0ccf691a.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/zby_ztmjczc.mp3"
+dest_files=["res://.godot/imported/zby_ztmjczc.mp3-849bbce90b59ebabfadd754a0ccf691a.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Assets/training_speech/zbzz_ckcjwb.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://b5wej2w00ttnj"
+path="res://.godot/imported/zbzz_ckcjwb.mp3-3c85dcaed29528587c7e0c8e32ccee6b.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/zbzz_ckcjwb.mp3"
+dest_files=["res://.godot/imported/zbzz_ckcjwb.mp3-3c85dcaed29528587c7e0c8e32ccee6b.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Assets/training_speech/zlgzmhfzc.mp3.import

@@ -0,0 +1,19 @@
+[remap]
+
+importer="mp3"
+type="AudioStreamMP3"
+uid="uid://bv1xfsb4450r1"
+path="res://.godot/imported/zlgzmhfzc.mp3-02c91e6200e7ec22dcd1900223f9e0da.mp3str"
+
+[deps]
+
+source_file="res://Assets/training_speech/zlgzmhfzc.mp3"
+dest_files=["res://.godot/imported/zlgzmhfzc.mp3-02c91e6200e7ec22dcd1900223f9e0da.mp3str"]
+
+[params]
+
+loop=false
+loop_offset=0
+bpm=0
+beat_count=0
+bar_beats=4

Communication/Assets/shoudao.mp3.import

@@ -1,19 +0,0 @@
-[remap]
-
-importer="mp3"
-type="AudioStreamMP3"
-uid="uid://dvvpoqb8umdjy"
-path="res://.godot/imported/shoudao.mp3-8e2429a197cf2b3306a67088ca44c2bf.mp3str"
-
-[deps]
-
-source_file="res://Communication/Assets/shoudao.mp3"
-dest_files=["res://.godot/imported/shoudao.mp3-8e2429a197cf2b3306a67088ca44c2bf.mp3str"]
-
-[params]
-
-loop=false
-loop_offset=0
-bpm=0
-beat_count=0
-bar_beats=4

Communication/Assets/zhenglieguzhangmenhuifuzhengchang.mp3.import

@@ -1,19 +0,0 @@
-[remap]
-
-importer="mp3"
-type="AudioStreamMP3"
-uid="uid://bv1xfsb4450r1"
-path="res://.godot/imported/zhenglieguzhangmenhuifuzhengchang.mp3-e699f05501af2fc615013fe553302b4f.mp3str"
-
-[deps]
-
-source_file="res://Communication/Assets/zhenglieguzhangmenhuifuzhengchang.mp3"
-dest_files=["res://.godot/imported/zhenglieguzhangmenhuifuzhengchang.mp3-e699f05501af2fc615013fe553302b4f.mp3str"]
-
-[params]
-
-loop=false
-loop_offset=0
-bpm=0
-beat_count=0
-bar_beats=4

Communication/voice_communication.gd

@@ -1,8 +1,9 @@
 extends Node
 
+## 录音效果器
 var effect: AudioEffectRecord
-var recording: AudioStreamWAV
-var record_stream: AudioStreamWAV
+## 录音捕获效果器（用于判断录音音量）
+var capture: AudioEffectCapture
 
 
 func _ready():
@@ -11,9 +12,7 @@ func _ready():
 	# And use it to retrieve its first effect, which has been defined
 	# as an "AudioEffectRecord" resource.
 	effect = AudioServer.get_bus_effect(idx, 0)
-	record_stream = AudioStreamWAV.new()
-	startRecord()
-	
+	capture = AudioServer.get_bus_effect(idx, 1)
 
 ## 启动录音
 func startRecord():
@@ -26,13 +25,38 @@ func stopRecord():
 	if effect.is_recording_active():
 		effect.set_recording_active(false)
 
-var audio_sd = preload("res://Communication/Assets/shoudao.mp3")
-var audio_zlgzmhfzc = preload("res://Communication/Assets/zhenglieguzhangmenhuifuzhengchang.mp3")
+var audio_sd = preload("res://Assets/training_speech/sd.mp3")
+var audio_zlgzmhfzc = preload("res://Assets/training_speech/zlgzmhfzc.mp3")
+
+
+## 播放回复
+## PS: 是协程函数，外部可以await
+func play_reply(reply):
+	stopRecord()
+	assert(reply is AudioStream, "reply不是音频资源")
+	## 确保不循环播放
+	if reply is AudioStreamMP3:
+		reply.loop = false
+	if reply is AudioStreamOggVorbis:
+		reply.loop = false
+	if reply is AudioStreamWAV:
+		reply.loop_mode = AudioStreamWAV.LOOP_DISABLED
+	$AudioStreamPlayer.stream = reply
+	$AudioStreamPlayer.play()
+	await $AudioStreamPlayer.finished
+
+## 录音并语音识别检查
+## PS: 是协程函数，外部如果关心结果需await
+func speech_record_check(keywords: Array[String]):
+	pass
 
 func _on_timer_timeout():
 	if effect.is_recording_active():
-		var vdb = $AudioStreamRecord.get_volume_db()
-		print(vdb)
+		print(capture.get_buffer_length_frames(), ", ", capture.get_discarded_frames(), ", ", capture.get_frames_available())
+		var buf = capture.get_buffer(capture.get_frames_available())
+		for vec in buf:
+			if vec.x > 0.1 or vec.y > 0.1:
+				print("Left channel volume = ", vec.x, ", Right volume = ", vec.y)
 	#if effect.is_recording_active():
 		#recording = effect.get_recording()
 		#print(recording.data.size(), ", ", recording.format, ", ", recording.loop_mode, ", ", recording.get_length())

Communication/voice_communication.tscn

@@ -16,7 +16,7 @@ bus = &"Record"
 autoplay = true
 
 [node name="Timer" type="Timer" parent="."]
-wait_time = 0.1
+wait_time = 0.2
 autostart = true
 
 [connection signal="timeout" from="Timer" to="." method="_on_timer_timeout"]

default_bus_layout.tres

@@ -1,8 +1,12 @@
-[gd_resource type="AudioBusLayout" load_steps=2 format=3 uid="uid://b4m2yc5k0re6f"]
+[gd_resource type="AudioBusLayout" load_steps=3 format=3 uid="uid://b4m2yc5k0re6f"]
 
 [sub_resource type="AudioEffectRecord" id="AudioEffectRecord_bjx8q"]
 resource_name = "Record"
 
+[sub_resource type="AudioEffectCapture" id="AudioEffectCapture_fxogc"]
+resource_name = "Capture"
+buffer_length = 1.0
+
 [resource]
 bus/1/name = &"Record"
 bus/1/solo = false
@@ -12,3 +16,5 @@ bus/1/volume_db = 0.0
 bus/1/send = &"Master"
 bus/1/effect/0/effect = SubResource("AudioEffectRecord_bjx8q")
 bus/1/effect/0/enabled = true
+bus/1/effect/1/effect = SubResource("AudioEffectCapture_fxogc")
+bus/1/effect/1/enabled = true

export_presets.cfg

@@ -1,8 +1,8 @@
 [preset.0]
 
-name="station_keys"
+name="Web"
 platform="Web"
-runnable=false
+runnable=true
 dedicated_server=false
 custom_features=""
 export_filter="exclude"
@@ -22,49 +22,7 @@ custom_template/release=""
 variant/extensions_support=false
 vram_texture_compression/for_desktop=true
 vram_texture_compression/for_mobile=false
-html/export_icon=true
-html/custom_html_shell=""
-html/head_include="<style>
-body {
-	overflow: hidden;
-}
-</style>"
-html/canvas_resize_policy=2
-html/focus_canvas_on_start=true
-html/experimental_virtual_keyboard=false
-progressive_web_app/enabled=false
-progressive_web_app/offline_page=""
-progressive_web_app/display=1
-progressive_web_app/orientation=0
-progressive_web_app/icon_144x144=""
-progressive_web_app/icon_180x180=""
-progressive_web_app/icon_512x512=""
-progressive_web_app/background_color=Color(0, 0, 0, 1)
-
-[preset.1]
-
-name="Web"
-platform="Web"
-runnable=true
-dedicated_server=false
-custom_features=""
-export_filter="all_resources"
-include_filter=""
-exclude_filter=""
-export_path="../export/total.html"
-encryption_include_filters=""
-encryption_exclude_filters=""
-encrypt_pck=false
-encrypt_directory=false
-
-[preset.1.options]
-
-custom_template/debug=""
-custom_template/release=""
-variant/extensions_support=false
-vram_texture_compression/for_desktop=true
-vram_texture_compression/for_mobile=false
-html/export_icon=true
+html/export_icon=false
 html/custom_html_shell=""
 html/head_include=""
 html/canvas_resize_policy=2

trainings/training1.gd

@@ -5,10 +5,16 @@ extends Node
 ## 场景描述
 @export var trainingDesc: String = "场景描述"
 
+var a_sd = preload("res://Assets/training_speech/sd.mp3")
+var a_xt_pslgbztmwx = preload("res://Assets/training_speech/xt_pslgbztmwx.mp3")
+var a_zby_ztmjczc = preload("res://Assets/training_speech/zby_ztmjczc.mp3")
+
 func _ready():
 	$TrainingDescDialog.title = trainingName
 	$TrainingDescDialog.updateSceneDesc(trainingDesc)
 
 
 func _on_training_desc_dialog_start_training():
-	pass # Replace with function body.
+	await $VoiceCommunication.play_reply(a_xt_pslgbztmwx)
+	await $VoiceCommunication.play_reply(a_zby_ztmjczc)
+	await $VoiceCommunication.play_reply(a_sd)

trainings/training1.tscn

@@ -1,7 +1,8 @@
-[gd_scene load_steps=3 format=3 uid="uid://dvsauaycyah6j"]
+[gd_scene load_steps=4 format=3 uid="uid://dvsauaycyah6j"]
 
 [ext_resource type="Script" path="res://trainings/training1.gd" id="1_71wgx"]
 [ext_resource type="PackedScene" uid="uid://qh8b3g6n3k2u" path="res://trainings/training_desc_dialog.tscn" id="2_oc00d"]
+[ext_resource type="PackedScene" uid="uid://6lortgihx5jy" path="res://Communication/voice_communication.tscn" id="3_ceo1a"]
 
 [node name="Training1" type="Node"]
 script = ExtResource("1_71wgx")
@@ -11,4 +12,6 @@ trainingDesc = "列车关门作业，突发单个站台门未正常关闭。"
 [node name="TrainingDescDialog" parent="." instance=ExtResource("2_oc00d")]
 size = Vector2i(400, 300)
 
+[node name="VoiceCommunication" parent="." instance=ExtResource("3_ceo1a")]
+
 [connection signal="StartTraining" from="TrainingDescDialog" to="." method="_on_training_desc_dialog_start_training"]