Speech Ability

Introduction

Speech Ability includes listening and speaking abilities.

Ability to listen:

Ability to speak:

Create speech service callback

SkillCallback mSkillCallback = new SkillCallback() {    
    @Override    
    public void onSpeechParResult(String s) throws RemoteException {        
        //The result of temporary speech recognition    
    }    
    @Override    
    public void onStart() throws RemoteException {        
        //Start recognition    
    }    
    @Override    
    public void onStop() throws RemoteException {        
        // end of recognition    
    }    
    @Override    
    public void onVolumeChange(int volume) throws RemoteException {        
        //The size of the recognized voice changes    
    }    
    /**     
     * @param status 0: return normally     
     * 1: other returns     
     * 2: Noise or single_other return     
     * 3: Timeout     
     * 4: Forced to cancel     
     * 5: The asr result ends early, without passing through NLU     
     * 6: In the case of full duplex with the same semantics, other returns     
     */    
    @Override    
    public void onQueryEnded(int status) throws RemoteException {    
    }    
    @Override    
    public void onQueryAsrResult(String asrResult) throws RemoteException {        
        //asrResult: final recognition result    
    }
};

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Connect to speech service

final SkillApi skillApi = new SkillApi();
skillApi.connectApi(context, new ApiListener() {     
    @Override     
    public void handleApiDisabled() {     
    }     
    @Override     
    public void handleApiConnected() {         
        //Voice service connection is successful, register voice callback                 
        skillApi.registerCallBack(mSkillCallback);     
    }     
    @Override     
    public void handleApiDisconnected() {         
        //Voice service has been disconnected     
    }
});

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

TTS playback

Method name: playText

Call method：

skillApi.playText(text, new TextListener() {     
    @Override     
    public void onStart() {         
        //Play start     
    }     
    @Override     
    public void onStop() {         
        //Play stop     
    }     
    @Override     
    public void onError() {         
        //Play error     
    }     
    @Override     
    public void onComplete() {         
        //Play is complete     
    }
});

Parameter Description:

• text : TTS playback text Note: Please avoid using long strings in TTS text that we cannot support, such as hyperlinks, scrambled character sequences, etc. This may cause program calls to fail. It is recommended to keep a single TTS playback within 300 characters, with a maximum limit of 1000 characters. If the text content is too long, it can be divided into segments for TTS playback.

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Stop TTS playback

Method name: stopTTS

Calling method:

skillApi.stopTTS();

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Set speech recognition mode

Note: The interface will only take effect when “Voice Settings” and “Continuous Pickup” are turned on in the robot settings

Method name: setRecognizeMode

Calling method:

skillApi.setRecognizeMode(isContinue);

Parameter Description:

• isContinue: boolean type, true continuous recognition false single recognition

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Turn off or turn on speech recognition

Note: The interface will only take effect when “Voice Settings” and “Continuous Pickup” are turned on in the robot settings

Method name: setRecognizable

Calling method:

skillApi.setRecognizable(enable);

Parameter Description:

• enable: boolean type, true open false close

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Set speech recognition area

Method name: setAngleCenterRange

Calling method:

skillApi.setAngleCenterRange(angleCenter, angleRange);

Only the first generation algorithm supports setting the center angle for this interface, while the second generation algorithm does not.

Parameter Description:

• angleCenter : center angle, float type, [0,360)
• angleRange : interval angle, float type, 0,120

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	NO	NO	NO	Yes

Get NLP results through text

Method name: queryByText

Calling method:

skillApi.queryByText(text);

Parameter Description:

• text : the text to be queried, String type

Note: This interface has no return value. The result of the query is returned as a voice command through onSendRequest of ModuleCallback.

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

Sound recording

This is a part about sound recording.

//Audio sampling rate (fixed 16000)
private final int AUDIO_RATE = 16000;
//Buffer length
private final int AUDIO_BUF_LEN = 640;
//Channel (stereo)
int channelConfig = AudioFormat.CHANNEL_IN_STEREO;
int minRecBufSize = AudioRecord.getMinBufferSize(AUDIO_RATE, channelConfig, AudioFormat.ENCODING_PCM_16BIT);
int recBufSize = minRecBufSize * 2;
AudioRecord audioRecorder =new AudioRecord(AudioSource.MIC, AUDIO_RATE, AudioFormat.CHANNEL_IN_STEREO, AudioFormat.ENCODING_PCM_16BIT, recBufSize);
//data collection
byte[] tempBufRec = new byte[AUDIO_BUF_LEN];
byte[] bfOutLeft = new byte[AUDIO_BUF_LEN / 2];
byte[] bfOutRight = new byte[AUDIO_BUF_LEN / 2];
int bufferSize = 5 * 2 * 16000;
int readBytes = audioRecorder.read(tempBufRec, 0, bufferSize);
if (readBytes == bufferSize) {    
    deinterleaveData(tempBufRec, bfOutLeft, bfOutRight, bufferSize);    
    //The separated data, bfOutLeft is the final required data
}
//Left and right channel data separation
private void deinterleaveData(byte[] src, byte[] leftdest, byte[] rightdest, int len) {    
    int rIndex = 0;    
    int lIndex = 0;    
    for (int i = 0; i <len;) {        
        leftdest[rIndex] = src[i];        
        leftdest[rIndex + 1] = src[i + 1];        
        rIndex = rIndex + 2;        
        rightdest[lIndex] = src[i + 2];        
        rightdest[lIndex + 1] = src[i + 3];        
        lIndex = lIndex + 2;        
        i = i + 4;    
    }
}

Note1: All parameters are fixed parameters and cannot be changed, otherwise the data may not be collected.

Note2: All the robots don’t support VOIP/SIP protocol, even if use those settings.

There is a demo to show this function(worked with Mini): Download

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes

text to mp3

Method name: textToMp3 Notice: this API only work with Chinese Launguage.

Calling method:

boolean status = RobotApi.getInstance().textToMp3(reqId, text, path, name,    new CommandListener(){        
    @Override        
    public void onResult(int result, String message) {            
        if (Definition.RESULT_OK == result) {                
            try {                    
                JSONObject jsonObject = new JSONObject(message);                    
                int code = jsonObject.optInt("code");                    
                String errMessage = jsonObject.optString("message");                    
                if (code == 0) {                        
                    mediaPlayer = new MediaPlayer();                        
                    mediaPlayer.setDataSource(path + File.separator + name);                                
                    mediaPlayer.prepare();                        
                    mediaPlayer.start();                    
                } else{                    
                }                
            } catch (JSONException | IOException e) {                    
                e.printStackTrace();                
            }            
        }else {            
        }        
    }  
});

Parameter Description:

• reqId: int type command id, pass 0
• text: String type to be converted text
• path: String type file path
• name: String type file name
• listener: CommandListener message callback{“code”:0, “message”:”err msg”}

Return Value:

int result 0 command executed / -1 not executed

Note:

1. To call the interface, you need the app to apply for sdcard usage rights

2. After the conversion is successful, an mp

3 file will be generated/overwritten in the specified path 3. Please manage the file yourself

Applicable Platform:

GreetBot	Mini	Lucki	DeliverBot	BigScreenBot
Yes	Yes	Yes	Yes	Yes