Fun-ASR Real-Time Speech Recognition Android SDK

Call Steps

1. Initialize the SDK.

2. Set parameters based on your business needs. Use the parameters parameter of the initialize method to set connection and control parameters. Use the setParams method to set speech recognition effect parameters.

3. Call startDialog to start the recognition process.

4. In the onNuiAudioStateChanged callback, enable the audio recording device based on the audio state.

5. In the onNuiNeedAudioData callback, continuously provide recorded audio data.

6. In the onNuiEventCallback callback, listen for events and retrieve speech recognition results.

7. Call stopDialog to stop the recognition. You can confirm that the recognition has ended by listening for the EVENT_TRANSCRIBER_COMPLETE event.

8. When you no longer need the recognition feature, call the release method to release SDK resources.

Connection and Control Parameters

You can configure these parameters by passing a JSON string through the parameters argument of the initialize method.

• Parameter example: The following code shows a sample JSON string. Not all parameters are listed. You can add other parameters as needed during implementation.

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

• Parameter description

  Parameter	Type	Required	Description
  url	String	Yes	Service endpoint. Fixed as wss://dashscope-intl.aliyuncs.com/api-ws/v1/inference.
  apikey	String	Yes	API key.
  service_mode	String	Yes	Operation mode. For real-time speech recognition, this must be "1".
  device_id	String	Yes	A unique string identifying the end user. Set this to an in-app user ID or a client-generated device identifier. This ID is used primarily for log tracing and troubleshooting.
  debug_path	String	No	Storage path for log files. This parameter takes effect only when save_log is set to true in the initialize call. You must specify a log file path in this case, or an error occurs. The system keeps at most two local log files.
  save_wav	String	No	Whether to save audio files for debugging. Audio files are saved under the debug_path directory. Default value: "false". Valid values: "true": yes "false": no This parameter takes effect only when save_log is set to true in the initialize call. You must also set debug_path.
  max_log_file_size	int	No	Maximum size (in bytes) for each log file. This parameter takes effect only when save_log is set to true in the initialize call. Default value: 104857600 (100 × 1024 × 1024 bytes, or 100 MiB).
  log_track_level	int	No	Controls the filter level of log content sent to an external destination through the onNuiLogTrackCallback callback. Default value: 2. Valid values: 0: LOG_LEVEL_VERBOSE 1: LOG_LEVEL_DEBUG 2: LOG_LEVEL_INFO 3: LOG_LEVEL_WARNING 4: LOG_LEVEL_ERROR 5: LOG_LEVEL_NONE (disables this feature) Note: log_track_level and level (set via the initialize interface) jointly determine which logs are finally sent through the callback. A log entry is sent only if its level number is greater than or equal to both log_track_level and level. For example, if log_track_level is set to 2 (INFO) and level is set to 3 (WARNING), only logs at WARNING level or higher (numeric value ≥ 3) are sent.

Speech Recognition Effect Parameters

You can configure these parameters by passing a JSON string through the params argument of the setParams method.

• Parameter example: The following code shows a sample JSON string. Not all parameters are listed. You can add other parameters as needed during implementation.

  {
      "service_type": 4,
      "nls_config": {
          "model": "fun-asr-realtime",
          "sr_format": "pcm",
          "sample_rate": 16000,
          "parameters": {
              "speech_noise_threshold": 0.0
          }
      }
  }

• Parameter description

  Top-level parameter	Type	Required	Description
  service_type	int	Yes	Voice service type. For real-time speech recognition, this must be 4.
  nls_config	object	Yes	Core speech recognition configuration object containing key parameters such as model selection and recognition behavior control.
  nls_config.model	string	Yes	Speech recognition model.
  nls_config.sr_format	string	Yes	Audio format for recognition. Supported formats: pcm, wav, opus. Important opus: Must be PCM-encoded. The SDK internally encodes it into OPUS format. wav/pcm: Must be PCM-encoded.
  nls_config.sample_rate	int	Yes	Audio sampling rate in Hz. Only 16000 Hz is supported.
  nls_config.semantic_punctuation_enabled	boolean	No	Sets sentence segmentation mode. Default value: false. Valid values: true: Enables semantic punctuation and disables VAD (Voice Activity Detection) segmentation. false: Enables VAD segmentation and disables semantic punctuation. Semantic punctuation offers higher accuracy and suits meeting transcription scenarios. VAD segmentation has lower latency and suits real-time interactive scenarios.
  nls_config.max_sentence_silence	int	No	VAD (Voice Activity Detection) silence threshold in milliseconds for sentence segmentation. Default value: 800. Valid range: [200, 6000]. When silence after speech exceeds this threshold, the system considers the sentence complete. This parameter takes effect only when semantic_punctuation_enabled is false.
  nls_config.multi_threshold_mode_enabled	boolean	No	Enables protection against overly long VAD segmentation. When enabled, it prevents VAD from cutting sentences too long. Default value: false (disabled). Valid values: true: enabled false: disabled This parameter takes effect only when semantic_punctuation_enabled is false.
  nls_config.heartbeat	boolean	No	Is a persistent connection to the server-side maintained? Default value: false. Valid values: true: Keeps the connection alive even when sending continuous silence audio. false: Closes the connection after 60 seconds of inactivity due to timeout. This 60-second timeout is a server-side default and cannot be configured on the client.
  nls_config.vocabulary_id	string	No	ID of a hotword vocabulary list to improve recognition accuracy for specific terms. For instructions on using hotwords, see Customize Hotwords.
  nls_config.language_hints	array[string]	No	Sets the language codes for recognition. If the language is unknown in advance, leave this parameter unset and the model will identify it automatically. The system reads only the first value in the array and ignores all other values. Supported language codes by model: fun-asr-realtime, fun-asr-realtime-2025-11-07: zh: Chinese en: English ja: Japanese fun-asr-realtime-2025-09-15: zh: Chinese en: English
  nls_config.parameters	object	No	Configures additional parameters as a JSON object.
  nls_config.parameters.speech_noise_threshold	float	No	Adjusts the speech-noise detection threshold to control VAD sensitivity. Range: [-1.0, 1.0]. Guidelines: Near -1: Lowers the noise threshold — more noise may be transcribed as speech. Near +1: Raises the noise threshold — some speech may be filtered out as noise. Important This is an advanced parameter. Adjustments can significantly affect recognition quality. Test thoroughly before adjusting. Make small adjustments (step size 0.1) based on your audio environment.

NativeNui

INativeNuiCallback: Listener Callbacks

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

NuiEvent: Event Types