public void StopRecord()
{
if (!IsRecording || !ReadyToRecord())
{
return;
}
IsRecording = false;
CustomMicrophone.End(MicrophoneDevice);
if (!DetectVoice)
{
LastRecordedRaw = _currentRecordingVoice.ToArray();
LastRecordedClip = AudioConvert.Convert(LastRecordedRaw, _microphoneWorkingAudioClip.channels);
}
if (_currentRecordingVoice != null)
{
_currentRecordingVoice.Clear();
}
_currentAudioSamples = null;
_currentRecordingVoice = null;
#if NET_2_0 || NET_2_0_SUBSET
if (RecordEndedEvent != null)
{
RecordEndedEvent(LatestVoiceAudioClip, LastRecordedRaw);
}
#else
RecordEndedEvent?.Invoke(LastRecordedClip, LastRecordedRaw);
#endif
}
public void Recognize(AudioClip clip, List <string[]> contexts, Enumerators.LanguageCode language)
{
if (_currentConfig == null)
{
throw new NotImplementedException("Config isn't seted! Use SetConfig method!");
}
if (clip == null)
{
throw new NotImplementedException("AudioClip isn't seted!");
}
string postData = string.Empty;
string uri = string.Empty;
switch (_currentConfig.recognitionType)
{
case Enumerators.GoogleNetworkType.SPEECH_RECOGNIZE:
{
if (!_gcSpeechRecognition.isUseAPIKeyFromPrefab)
{
uri = Constants.RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + Constants.GC_API_KEY;
}
else
{
uri = Constants.RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + _gcSpeechRecognition.apiKey;
}
postData = JsonUtility.ToJson(GenerateRecognizeRequest(
AudioConvert.Convert(clip, _currentConfig.audioEncoding,
_currentConfig.useVolumeMultiplier,
_currentConfig.audioVolumeMultiplier), contexts, language));
}
break;
case Enumerators.GoogleNetworkType.SPEECH_LONGRECOGNIZE:
{
if (!_gcSpeechRecognition.isUseAPIKeyFromPrefab)
{
uri = Constants.LONG_RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + Constants.GC_API_KEY;
}
else
{
uri = Constants.LONG_RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + _gcSpeechRecognition.apiKey;
}
postData = JsonUtility.ToJson(GenerateRecognizeRequest(
AudioConvert.Convert(clip, _currentConfig.audioEncoding,
_currentConfig.useVolumeMultiplier,
_currentConfig.audioVolumeMultiplier), contexts, language));
}
break;
default:
throw new NotSupportedException(_currentConfig.recognitionType + " doesn't supported!");
}
_networking.SendRequest(uri, postData, _currentConfig.recognitionType);
}
public void Recognize(AudioClip clip, List <string[]> contexts, Enumerators.LanguageCode language)
{
if (_currentConfig == null)
{
throw new NotImplementedException("Config isn't seted! Use SetConfig method!");
}
if (clip == null)
{
throw new NotImplementedException("AudioClip isn't seted!");
}
string postData = string.Empty;
string uri = string.Empty;
switch (_currentConfig.recognitionType)
{
case Enumerators.SpeechRecognitionType.SYNC:
{
uri = Constants.RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + Constants.GC_API_KEY;
postData = JsonUtility.ToJson(GenerateSyncRequest(
AudioConvert.Convert(clip, _currentConfig.audioEncoding,
_currentConfig.useVolumeMultiplier,
_currentConfig.audioVolumeMultiplier), contexts, language));
}
break;
case Enumerators.SpeechRecognitionType.ASYNC:
{
Debug.Log("Async(Long) speech recognition isn't fully implemented!");
uri = Constants.LONG_RECOGNIZE_REQUEST_URL + Constants.API_KEY_PARAM + Constants.GC_API_KEY;
postData = JsonUtility.ToJson(GenerateSyncRequest(
AudioConvert.Convert(clip, _currentConfig.audioEncoding,
_currentConfig.useVolumeMultiplier,
_currentConfig.audioVolumeMultiplier), contexts, language));
}
break;
default:
throw new NotSupportedException(_currentConfig.recognitionType + " doesn't supported!");
}
_networking.SendRequest(uri, postData);
}
public void Update()
{
if (IsRecording)
{
_currentSamplePosition = CustomMicrophone.GetPosition(MicrophoneDevice);
CustomMicrophone.GetRawData(ref _currentAudioSamples, _microphoneWorkingAudioClip);
if (DetectVoice)
{
bool isTalking = _voiceDetectionManager.HasDetectedVoice(AudioClip2ByteConverter.FloatToByte(_currentAudioSamples));
if (isTalking)
{
_endTalkingDelay = 0f;
}
else
{
_endTalkingDelay += Time.deltaTime;
}
if (!_isTalking && isTalking)
{
_isTalking = true;
#if NET_2_0 || NET_2_0_SUBSET
if (TalkBeganEvent != null)
{
TalkBeganEvent();
}
#else
TalkBeganEvent?.Invoke();
#endif
}
else if (_isTalking && !isTalking && _endTalkingDelay >= _speechRecognitionManager.CurrentConfig.voiceDetectionEndTalkingDelay)
{
_isTalking = false;
LastRecordedRaw = _currentRecordingVoice.ToArray();
LastRecordedClip = AudioConvert.Convert(LastRecordedRaw, _microphoneWorkingAudioClip.channels);
_currentRecordingVoice.Clear();
#if NET_2_0 || NET_2_0_SUBSET
if (TalkEnded != null)
{
TalkEnded(LatestVoiceAudioClip, LastRecordedRaw);
}
#else
TalkEndedEvent?.Invoke(LastRecordedClip, LastRecordedRaw);
#endif
}
else if (_isTalking && isTalking)
{
AddAudioSamplesIntoBuffer();
}
}
else
{
AddAudioSamplesIntoBuffer();
}
_previousSamplePosition = _currentSamplePosition;
}
}
