Use the HarmonyOS SDK for Qwen-Audio-3.0-Realtime to build real-time voice conversations with audio input and audio or text output.
User guide: For model introductions and selection advice, see Real-time voice conversation.
Unlike Android, which uses
Declare the microphone permission in
Pass a JSON string in the
Parameters
Pass a JSON string in the
Parameters
Initializes a voice conversation SDK instance. The SDK uses the singleton pattern. Do not initialize it again before you call release.
This method can block. Call it on a non-UI thread.
Method signature
Parameters
Return value
An error code. See Error code reference.
Sets the voice conversation parameters in JSON format. Call this method before startDialog.
Method signature
Parameters
Return value
An error code. See Error code reference.
Starts the conversation.
Method signature
Parameters
Return value
An error code. See Error code reference.
Ends the conversation. After you call this method, the server returns the final conversation result and ends the task.
Method signature
Return value
An error code. See Error code reference.
Ends the conversation immediately. After you call this method, the task ends without waiting for the server to return the final conversation result.
Method signature
Return value
An error code. See Error code reference.
Sends a conversation action command during an interaction to update runtime behavior such as the conversation context.
Method signature
Parameters
Example:
Return value
An error code. See Error code reference.
When
Parameters
Return value
An error code. See Error code reference.
When
Parameters
Return value
An error code. See Error code reference.
Releases all internal SDK resources. The SDK instance becomes unavailable after this call. To use the SDK again, call initialize to reinitialize it.
Method signature
Return value
An error code. See Error code reference.
Returns information about the current SDK version.
Method signature
Return value
Information about the current SDK version.
Method signature
Parameters
The SDK uses this callback to notify the application when to start or stop recording.
Method signature
AudioState values
Listens for the recording volume, which can be displayed in the UI.
Method signature
Parameters
After the conversation starts, this callback is triggered continuously. Supply the audio data to process in this callback. You do not need to use this callback when
Parameters
Return value
The actual number of bytes supplied.
Receives internal SDK auxiliary events and related data.
Method signature
Parameters
Receives detailed internal SDK logs for troubleshooting and debugging.
In the HarmonyOS SDK, event types are defined by the
Quick start
- Obtain and configure an API key.
-
Download the SDK and run the sample code:
- Download the latest SDK package.
- Extract the
.tar.gzSDK package. Obtain the HAR SDK fromentry/libsand add it to your project dependencies. For C++ integration, usenative/libsandnative/includein the package to obtain the dynamic libraries and header files. - Open the project in DevEco Studio. The sample code is in
DashQwenAudioChatPage.ets. Replace the API key to try the feature.
Call procedure
- Initialize the SDK.
- Set parameters for your use case. Use the
parametersargument of initialize to set the connection and control parameters, and use setParams to set the voice conversation parameters. - Call startDialog to start the conversation.
- In onNuiAudioStateChanged, start the recording device based on the audio state.
- Continuously supply recording data in onNuiNeedAudioData, or call updateAudio to actively push recording data.
- Continuously receive the audio returned by the model in onNuiAssistEventCallback.
- Listen for events and obtain event information in onNuiEventCallback.
- Call stopDialog to stop the conversation, and listen for
EVENT_TRANSCRIBER_COMPLETEto confirm that it has ended. - When the conversation feature is no longer needed, call release to release the SDK resources.
Audio device management
Unlike Android, which uses AudioRecord and AudioTrack, HarmonyOS provides audio capture and playback through @kit.AudioKit, using AudioCapturer for recording and AudioRenderer for playback. The sample project provides the AudioRecorder.ets and AudioPlayer.ets utility classes for reuse.
Recording (AudioCapturer)
- Create: Call
audio.createAudioCapturer(capturerOptions)asynchronously. The sample uses 16 kHz, 16-bit, mono audio (SAMPLE_RATE_16000,CHANNEL_1,SAMPLE_FORMAT_S16LE, andENCODING_TYPE_RAW). -
Audio source (
audio.SourceType):SOURCE_TYPE_MIC: Raw microphone audio. Use this source when SDK-internal AEC is enabled, and send data to the SDK by callingupdateAudio.SOURCE_TYPE_VOICE_COMMUNICATION: Call audio on which the system has already performed echo cancellation. Use this source when SDK-internal AEC is disabled, and provide audio throughonNuiNeedAudioData.
- Data event: Call
capturer.on('readData', (buffer: ArrayBuffer) => void)to continuously obtain recorded audio. - State event: Call
capturer.on('stateChange', (state: audio.AudioState) => void).STATE_RUNNINGindicates that recording has started, andSTATE_STOPPEDindicates that it has stopped. - Control: Call
start()to start,stop()to stop, andrelease()to release the recorder.
Note: HarmonyOS createsAudioCapturerasynchronously. Callstart()only after creation is complete. Do not create and immediately start a recorder inSTATE_OPEN. Create it first, then callstart()from theSTATE_OPENcallback. The sample creates the recorder duringdoInitand starts it duringonNuiAudioStateChanged.
Playback (AudioRenderer)
- Create: Call
audio.createAudioRenderer(rendererOptions)asynchronously. - Sample rate: The synthesized response audio returned by DashScope Realtime is 24 kHz. Pass this rate to the
AudioPlayerconstructor. - Data event: Call
renderer.on('writeData', (data: ArrayBuffer): audio.AudioDataCallbackResult => ...)to provide audio for playback. ReturnAudioDataCallbackResult.VALIDafter filling the buffer orINVALIDwhen no data is available. - State event: Call
renderer.on('stateChange', (state: audio.AudioState) => void)to listen for playback start and end events. - Control: Call
start()to start,stop()to stop, andpause()to pause playback. Pausing retains buffered data.
AEC reference signal: When SDK-internal AEC is enabled, send the player's output audio to the SDK as the reference signal by callingnuiInstance.pushReferenceData(data, false), which corresponds toupdateRefAudioon Android.
Permission declaration
Declare the microphone permission in module.json5 before recording audio:
Request parameters
Connection and control parameters
Pass a JSON string in the parameters argument of initialize.
Example: The following JSON string does not list every parameter. Add parameters as needed for your use case.
| Parameter | Type | Required | Description |
|---|---|---|---|
url | String | Yes | Service endpoint:
{WorkspaceId} with your actual Workspace ID. |
apikey | String | Yes | API key. |
service_mode | String | Yes | Runtime mode. Set this parameter to "1" for real-time voice conversation. |
device_id | String | Yes | A unique string that identifies the end user. You can use an in-app user ID or a client-generated device identifier. This ID is mainly used for log tracing and troubleshooting. |
audio_update_manually | String | No | Whether to actively push audio data. Default: "false". If set to "true" and the SDK supports on-device audio capabilities such as AEC and VAD, those capabilities are enabled by default. |
workspace | String | No | The storage path for on-device resource files. This parameter is required when audio_update_manually is "true" and an on-device audio capability such as AEC or VAD is enabled. |
debug_path | String | No | The log file storage path. This parameter takes effect only when save_log is true in initialize. In this case, the path is required. The SDK retains at most two log files locally. |
save_wav | String | No | Whether to save debug audio under debug_path. Default: "false". Valid values are "true" and "false". This parameter takes effect only when save_log is true, and debug_path must also be set. |
max_log_file_size | int | No | The maximum log file size in bytes. This parameter takes effect only when save_log is true. Default: 104857600 (100 × 1024 × 1024 bytes, or 100 MiB). |
log_track_level | int | No | The filter level for logs sent through onNuiLogTrackCallback. Default: 2. Valid values: 0 (VERBOSE), 1 (DEBUG), 2 (INFO), 3 (WARNING), 4 (ERROR), and 5 (NONE). A log is returned only when its level is greater than or equal to both log_track_level and the level passed to initialize. For example, if log_track_level is 2 (INFO) and level is 3 (WARNING), only WARNING and higher-level logs (values greater than or equal to 3) are returned. |
aec_params | object | No | Advanced on-device AEC configuration. This object takes effect only when audio_update_manually is "true". |
aec_params.enable_aec | boolean | No | Whether to enable on-device AEC. If audio_update_manually is "true" and the SDK supports on-device AEC, AEC is enabled by default. |
aec_params.save_audio | boolean | No | Whether to save audio processed by the on-device AEC module. If save_wav is "true" and debug_path is set, this feature is enabled by default and the audio is saved under debug_path. |
aec_params.enable_aec_data_callback | boolean | No | Whether to return AEC-processed audio through EVENT_AEC_DATA in onNuiAssistEventCallback. Default: false. |
vad_params | object | No | Advanced on-device VAD configuration. This object takes effect only when audio_update_manually is "true". |
vad_params.enable_vad | boolean | No | Whether to enable on-device VAD. If audio_update_manually is "true" and the SDK supports on-device VAD, VAD is enabled by default. |
vad_params.save_audio | boolean | No | Whether to save audio processed by the on-device VAD module. If save_wav is "true" and debug_path is set, this feature is enabled by default and the audio is saved under debug_path. |
Voice conversation parameters
Pass a JSON string in the params argument of setParams.
Example: The following JSON string does not list every parameter. Add parameters as needed for your use case.
| Top-level parameter | Type | Required | Description |
|---|---|---|---|
service_type | int | Yes | Voice service type. Set this parameter to 4 for real-time voice conversation. |
nls_config | object | Yes | Core voice conversation configuration, including model selection and conversation behavior. |
nls_config.model | string | Yes | Model name. Supports the qwen-audio-3.1-realtime-plus, qwen-audio-3.0-realtime-plus, and qwen-audio-3.0-realtime-flash model series. |
nls_config.sr_format | string | Yes | Input audio format. Only pcm is supported. The default format is 16 kHz, 16-bit, mono PCM. |
nls_config.modalities | string | No | A string containing an array of output modalities. Valid values:
|
nls_config.voice | string | No | TTS voice. The default is longanqian_v3.1 for 3.1 Plus and longanqian for 3.0 Plus/Flash. This parameter can be set only in the first session.update; later values are ignored. System voices: longanqian, longanlingxin, longanlingxi, longanxiaoxin, and longanlufeng. You can also specify a cloned voice_id created through the voice cloning API. See Voice configuration.3.1 Plus also supports longanqian_v3.1, longanhuan_v3.1, longanlingxin_v3.1, longanfengyue_v3.1, xunanchuan_v3.1, beth_v3.1, betty_v3.1, cally_v3.1. |
nls_config.enable_speech_emotion | boolean | No | Whether to enable enhanced emotional expression. When enabled, the response voice has more noticeable emotional variation. Default: true. Valid values: true and false. |
nls_config.instructions | string | No | System instructions that define the model's role, response style, and behavioral preferences for the entire session. |
nls_config.max_history_turns | int | No | Maximum number of historical question-answer turns allowed in a request. Valid values: 1 to 50. Default: 20. |
nls_config.tools | string | No | A string containing an array of Function Calling tool definitions. After you configure this parameter, the model decides whether to call a tool based on the user input. Each definition uses type (required and fixed to function) and a function object containing name (required), description (optional), and parameters (optional). If a function has no arguments, omit parameters.Example: |
nls_config.turn_detection | string | No | A string containing a JSON object for turn detection. If omitted, the session uses push-to-talk mode, in which audio is committed and inference is triggered manually. If set, duplex conversation mode is enabled. |
nls_config.turn_detection.type | string | No | VAD type. server_vad (default) detects speech boundaries from acoustic features and automatically triggers inference. smart_turn combines acoustic and semantic signals; sounds without semantic content, such as filler sounds, do not start a turn or interrupt model playback. |
nls_config.turn_detection.threshold | float | No | VAD sensitivity. This parameter applies only to server_vad and has no effect on smart_turn. Lower values make VAD more sensitive to quiet sounds and background noise; higher values require clearer and louder speech. Valid range: [-1.0, 1.0]. Default: 0.5. |
nls_config.turn_detection.silence_duration_ms | int | No | Minimum silence duration after speech, in milliseconds, before the model response is triggered. This parameter applies only to server_vad and has no effect on smart_turn. Lower values reduce latency but can trigger on brief pauses. Valid range: [200, 6000]. Default: 800. Recommended for conversation: 400 to 800. |
nls_config.turn_detection.voiceprint_audio_urls | string | No | A string containing an array of publicly accessible prerecorded audio URLs for the target speaker. This parameter applies only to smart_turn. In duplex conversations, it helps the model focus on the target speaker and ignore other speakers and background noise. Up to five URLs are supported. Audio must be 16 kHz PCM or WAV. |
Key APIs
NativeNui
initialize
Initializes a voice conversation SDK instance. The SDK uses the singleton pattern. Do not initialize it again before you call release.
This method can block. Call it on a non-UI thread.
Method signature
| Parameter | Type | Description |
|---|---|---|
callback | INativeNuiCallback | An implementation of the event and data callback interface. |
parameters | string | A JSON string that contains authentication, connection, and debugging parameters. See Connection and control parameters. |
level | number | Controls the print level of the SDK's own logs. |
save_log | boolean | Whether to save logs locally. If set to true, specify the path by using debug_path in the connection and control parameters. You can also set the file size by using max_log_file_size. |
setParams
Sets the voice conversation parameters in JSON format. Call this method before startDialog.
Method signature
| Parameter | Type | Description |
|---|---|---|
params | string | Voice conversation parameters. |
startDialog
Starts the conversation.
Method signature
| Parameter | Type | Description |
|---|---|---|
vad_mode | Constants.VadMode | VAD mode. Fixed to Constants.VadMode.TYPE_P2T. |
dialog_params | string | If apikey in the connection and control parameters is a temporary API key, update it here after it expires.JSON format: |
stopDialog
Ends the conversation. After you call this method, the server returns the final conversation result and ends the task.
Method signature
cancelDialog
Ends the conversation immediately. After you call this method, the task ends without waiting for the server to return the final conversation result.
Method signature
dialogAction
Sends a conversation action command during an interaction to update runtime behavior such as the conversation context.
Method signature
| Parameter | Type | Description |
|---|---|---|
params | string | A JSON string used to update runtime behavior such as the conversation context. |
params.type | String | Set to "action". |
params.command | String | Runtime command. Valid values:
|
params.context | String | Function call request update. Used when command is "function_call". |
params.context.type | String | Event type. This parameter is required when command is "function_call". conversation.item.create inserts a conversation item for historical context, supplemental text, or a tool result. After sending conversation.item.create, use response.create to trigger another inference. |
params.context.item | object | Required when params.context.type is conversation.item.create. The conversation item to create. |
params.context.response | object | Optional when params.context.type is response.create. Overrides the session defaults for this inference. If omitted, the current session configuration is used. |
context.item parameters:
| Parameter | Type | Description |
|---|---|---|
id | String | Optional unique conversation item ID. If omitted, the server generates one. An error is returned if the specified ID already exists. |
type | String | Required item type. Valid values:
|
role | String | Required for message. Valid values: system, user, and assistant. |
content | array | Required for message. Each item contains a type and the associated data field. system supports input_text with text; user supports input_text with text and input_audio with Base64-encoded audio; assistant supports output_text with text. |
call_id | String | Required for function_call and function_call_output. The unique ID that associates a function call request with its result. |
name | String | Required for function_call. The name of the function to call. |
arguments | String | Required for function_call. Function arguments as a JSON string. |
output | String | Required for function_call_output. Tool execution result as a JSON string. |
context.response parameters:
| Parameter | Type | Description |
|---|---|---|
modalities | array | Output modalities. ["text"] returns text only. ["audio", "text"] (default) returns both audio and text. |
voice | string | Overrides the TTS voice for this inference. |
updateAudio
When audio_update_manually is set to "true", recording data is no longer supplied through onNuiNeedAudioData. Use this method to actively push the data instead.
Method signature
| Parameter | Type | Description |
|---|---|---|
data | ArrayBuffer | Audio data to push (PCM). |
first_pack | boolean | Whether this is the first audio packet. The SDK calculates the byte count from data.byteLength. |
pushReferenceData (corresponds to Android updateRefAudio)
When audio_update_manually is set to "true" and on-device AEC is enabled, use this method to push audio played by the player as the reference signal.
Method signature
| Parameter | Type | Description |
|---|---|---|
data | ArrayBuffer | Audio data to push (PCM). |
first_pack | boolean | Whether this is the first audio packet. The SDK calculates the byte count from data.byteLength. |
release
Releases all internal SDK resources. The SDK instance becomes unavailable after this call. To use the SDK again, call initialize to reinitialize it.
Method signature
GetVersion
Returns information about the current SDK version.
Method signature
INativeNuiCallback: listener callbacks
onNuiEventCallback: listen for event information
Method signature
| Parameter | Type | Description |
|---|---|---|
event | Constants.NuiEvent | Callback event. |
resultCode | number | Error code. This parameter is valid when EVENT_ASR_ERROR occurs. |
arg2 | number | Reserved parameter. |
asrResult | AsrResult | Speech recognition result. |
kwsResult | KwsResult | Voice wake-up result. You do not need to use this parameter. |
onNuiAudioStateChanged: listen for the audio state
The SDK uses this callback to notify the application when to start or stop recording.
Method signature
| State | Description |
|---|---|
STATE_OPEN | The interaction has started. You can open the recording device and start recording. |
STATE_PAUSE | The interaction has stopped. You can stop recording. |
STATE_CLOSE | The SDK instance has been released. You can completely close the recording device. |
onNuiAudioRMSChanged: listen for the recording volume
Listens for the recording volume, which can be displayed in the UI.
Method signature
| Parameter | Type | Description |
|---|---|---|
val | number | Recording volume. |
onNuiNeedAudioData: supply audio data
After the conversation starts, this callback is triggered continuously. Supply the audio data to process in this callback. You do not need to use this callback when audio_update_manually is set to "true".
Method signature
| Parameter | Type | Description |
|---|---|---|
buffer | ArrayBuffer | Audio data to supply. The SDK uses buffer.byteLength as the requested byte count. |
onNuiAssistEventCallback: receive auxiliary data and information
Receives internal SDK auxiliary events and related data.
Method signature
| Parameter | Type | Description |
|---|---|---|
event | Constants.NuiEvent | NuiEvent event. |
info | string | Additional information, usually a JSON string. |
infoLen | number | Length of the additional information. |
data | ArrayBuffer | Additional binary data, such as TTS audio returned by the model. |
Note: This callback is optional (?) on HarmonyOS. Omit it if you do not need the information.
onNuiLogTrackCallback: listen for trace logs
Receives detailed internal SDK logs for troubleshooting and debugging.
NuiEvent: event types
In the HarmonyOS SDK, event types are defined by the Constants.NuiEvent enumeration. The following table lists the events related to real-time voice conversation:
| Event | Description |
|---|---|
EVENT_TRANSCRIBER_STARTED | The task started successfully. |
EVENT_VAD_START | Triggered immediately after the task starts. This does not mean that the start of speech has been detected. |
EVENT_VAD_END | The end of speech was detected. |
EVENT_ASR_PARTIAL_RESULT | An intermediate speech recognition result. |
EVENT_ASR_RESULT | A complete speech recognition result. |
EVENT_ASR_ERROR | An error occurred during the voice conversation. |
EVENT_MIC_ERROR | Triggered when no audio data is received for two consecutive seconds. |
EVENT_SENTENCE_START | The start of a sentence was detected. |
EVENT_SENTENCE_END | The end of a sentence was detected and a complete recognition result was returned. |
EVENT_TRANSCRIBER_COMPLETE | The voice conversation ended. |
EVENT_AUDIO_TRANSCRIPTION | An incremental text transcript event for audio output. Transcript segments are returned in streaming mode. |
EVENT_AUDIO_TRANSCRIPTION_COMPLETED | The transcript for audio output is complete. |
EVENT_OTHER_RESULT | Other event information, such as a Function Calling result. |
EVENT_ASR_TTS_START | The model started returning TTS audio. |
EVENT_ASR_TTS_DATA | TTS audio returned by the model. |
EVENT_ASR_TTS_COMPLETE | The model finished returning TTS audio. |
EVENT_RESULT_TRANSLATED | An intermediate translation result. |
EVENT_RESULT_TRANSLATED_END | Translation result output is complete. |
EVENT_AEC_DATA | Audio data processed by AEC. |