Qwen-Audio-ASR-Streaming Real-Time Speech Recognition Android SDK

This guide shows you how to use the Android SDK for Qwen-Audio-ASR-Streaming real-time speech recognition to convert speech to text.

Quick start

1. Obtain an API key

2. Download the SDK and run the sample code:
   • Download the latest SDK package.
   • Extract the ZIP package. The AAR-format SDK is in the app/libs directory. Add it to your project dependencies. For Android C++ integration, use android_libs and android_include from the ZIP package to get the dynamic libraries and header files.
   • Open the project in Android Studio. The sample code is in DashFunAsrSpeechTranscriberActivity.java. Replace the API key to try out the feature.

Call procedure

1. Initialize the SDK.
2. Set the parameters for your use case. Use the parameters argument of the initialize method to set the Connection and control parameters, and use the setParams method to set the Speech recognition parameters.
3. Call startDialog to start recognition.
4. In the onNuiAudioStateChanged callback, start the recording device based on the audio state.
5. Continuously supply recording data in the onNuiNeedAudioData callback, or actively push recording data by calling updateAudio.
6. In the onNuiEventCallback callback, listen for events and get the speech recognition results.
7. Call stopDialog to stop recognition, and listen for the EVENT_TRANSCRIBER_COMPLETE event to confirm that recognition has ended.
8. When you no longer need recognition, call release to release the SDK resources.

Request parameters

Connection and control parameters

To configure these parameters, pass a JSON string in the parameters argument of the initialize method.

Example: The following is a JSON string example. Not all parameters are listed. Add others as needed when you write your code:

{
    "url": "wss://dashscope.aliyuncs.com/api-ws/v1/inference",
    "apikey": "st-****",
    "device_id": "my_device_id",
    "service_mode": "1"
}

Speech recognition parameters

To configure these parameters, pass a JSON string in the params argument of the setParams method.

Example: The following is a JSON string example. Not all parameters are listed. Add others as needed when you write your code:

{
    "service_type": 4,
    "nls_config": {
        "model": "qwen-audio-3.0-asr-flash-streaming",
        "sr_format": "pcm",
        "sample_rate": "16000"
    }
}

Key interfaces

NativeNui

initialize

Initializes a speech recognition SDK instance. The SDK is a singleton. Do not initialize it more than once before you call release.

This method blocks, so call it on a non-UI thread.

public synchronized int initialize(final INativeNuiCallback callback,
                                   String parameters,
                                   final Constants.LogLevel level,
                                   final boolean save_log)

setParams

Sets the Speech recognition parameters in JSON format. Call this method before startDialog.

public synchronized int setParams(String params)

startDialog

Starts recognition.

public synchronized int startDialog(VadMode vad_mode, String dialog_params)

{
  "apikey": "st-****",
  "input_context": [
    {
      "role": "user",
      "content": [
        {
          "text": "xxxxx",
          "type": "input_text"
        }
      ]
    }
  ]
}

stopDialog

Ends recognition. After you call this method, the server returns the final recognition result and ends the task.

public synchronized int stopDialog();

cancelDialog

Ends recognition immediately. After you call this method, the task ends at once without waiting for the server to return the final recognition result.

public synchronized int cancelDialog();

updateAction

Sends an action command during an interaction to update runtime behavior, such as recognition context.

public synchronized int updateAction(String params);

{
  "context": [
    {
      "role": "user",
      "content": [
        {
          "text": "xxx",
          "type": "input_text"
        }
      ]
    }
  ]
}

updateAudio

When audio_update_manually is set to "true", call this method to actively push recording data instead of supplying the data through onNuiNeedAudioData.

public synchronized int updateAudio(byte[] data, int len,
                                    boolean first_pack);

updateRefAudio

When audio_update_manually is set to "true" and on-device AEC is enabled, call this method to push the audio played by the player as the reference signal.

public synchronized int updateRefAudio(byte[] data, int len,
                                       boolean first_pack);

release

Releases all internal resources of the SDK. After you call this method, the SDK instance becomes unavailable. To use it again, you must reinitialize it by calling initialize.

public synchronized int release();

GetVersion

Gets the current SDK version information.

public synchronized String GetVersion();

The current SDK version information.

INativeNuiCallback: listener callbacks

onNuiEventCallback: listen for events and speech recognition results

void onNuiEventCallback(NuiEvent event, final int resultCode, final int arg2, KwsResult kwsResult, AsrResult asrResult);

onNuiAudioStateChanged: listen for the audio state

The SDK uses this callback to notify you when to start or stop recording.

void onNuiAudioStateChanged(AudioState state);

onNuiNeedAudioData: supply the audio data to recognize

After recognition starts, this callback is triggered continuously. Supply the audio data to recognize in this callback.

int onNuiNeedAudioData(byte[] buffer, int len);

The number of bytes actually filled.

onNuiAssistEventCallback: receive auxiliary events and data

This callback receives auxiliary events and related data from the SDK.

void onNuiAssistEventCallback_(int event, byte[] info, int info_len,
                               byte[] data);

onNuiLogTrackCallback: listen for trace logs

This callback receives detailed internal logs from the SDK to help with troubleshooting and debugging.

default void onNuiLogTrackCallback(Constants.LogLevel level, String log)

NuiEvent: event types