azure-ai-voicelive-java – skill Lập trình cho agent AI

Azure AI VoiceLive SDK for Java

Real-time, bidirectional voice conversations with AI assistants using WebSocket technology.

Installation

<dependency>
    <groupId>com.azure</groupId>
    <artifactId>azure-ai-voicelive</artifactId>
    <version>1.0.0-beta.2</version>
</dependency>

Environment Variables

AZURE_VOICELIVE_ENDPOINT=https://<resource>.openai.azure.com/
AZURE_VOICELIVE_API_KEY=<your-api-key>

Authentication

API Key

import com.azure.ai.voicelive.VoiceLiveAsyncClient;
import com.azure.ai.voicelive.VoiceLiveClientBuilder;
import com.azure.core.credential.AzureKeyCredential;

VoiceLiveAsyncClient client = new VoiceLiveClientBuilder()
    .endpoint(System.getenv("AZURE_VOICELIVE_ENDPOINT"))
    .credential(new AzureKeyCredential(System.getenv("AZURE_VOICELIVE_API_KEY")))
    .buildAsyncClient();

DefaultAzureCredential (Recommended)

import com.azure.identity.DefaultAzureCredentialBuilder;

VoiceLiveAsyncClient client = new VoiceLiveClientBuilder()
    .endpoint(System.getenv("AZURE_VOICELIVE_ENDPOINT"))
    .credential(new DefaultAzureCredentialBuilder().build())
    .buildAsyncClient();

Key Concepts

| Concept | Description |

|---------|-------------|

| VoiceLiveAsyncClient | Main entry point for voice sessions |

| VoiceLiveSessionAsyncClient | Active WebSocket connection for streaming |

| VoiceLiveSessionOptions | Configuration for session behavior |

Audio Requirements

• Sample Rate: 24kHz (24000 Hz)
• Bit Depth: 16-bit PCM
• Channels: Mono (1 channel)
• Format: Signed PCM, little-endian

Core Workflow

1. Start Session

import reactor.core.publisher.Mono;

client.startSession("gpt-4o-realtime-preview")
    .flatMap(session -> {
        System.out.println("Session started");
        
        // Subscribe to events
        session.receiveEvents()
            .subscribe(
                event -> System.out.println("Event: " + event.getType()),
                error -> System.err.println("Error: " + error.getMessage())
            );
        
        return Mono.just(session);
    })
    .block();

2. Configure Session Options

import com.azure.ai.voicelive.models.*;
import java.util.Arrays;

ServerVadTurnDetection turnDetection = new ServerVadTurnDetection()
    .setThreshold(0.5)                    // Sensitivity (0.0-1.0)
    .setPrefixPaddingMs(300)              // Audio before speech
    .setSilenceDurationMs(500)            // Silence to end turn
    .setInterruptResponse(true)           // Allow interruptions
    .setAutoTruncate(true)
    .setCreateResponse(true);

AudioInputTranscriptionOptions transcription = new AudioInputTranscriptionOptions(
    AudioInputTranscriptionOptionsModel.WHISPER_1);

VoiceLiveSessionOptions options = new VoiceLiveSessionOptions()
    .setInstructions("You are a helpful AI voice assistant.")
    .setVoice(BinaryData.fromObject(new OpenAIVoice(OpenAIVoiceName.ALLOY)))
    .setModalities(Arrays.asList(InteractionModality.TEXT, InteractionModality.AUDIO))
    .setInputAudioFormat(InputAudioFormat.PCM16)
    .setOutputAudioFormat(OutputAudioFormat.PCM16)
    .setInputAudioSamplingRate(24000)
    .setInputAudioNoiseReduction(new AudioNoiseReduction(AudioNoiseReductionType.NEAR_FIELD))
    .setInputAudioEchoCancellation(new AudioEchoCancellation())
    .setInputAudioTranscription(transcription)
    .setTurnDetection(turnDetection);

// Send configuration
ClientEventSessionUpdate updateEvent = new ClientEventSessionUpdate(options);
session.sendEvent(updateEvent).subscribe();

3. Send Audio Input

byte[] audioData = readAudioChunk(); // Your PCM16 audio data
session.sendInputAudio(BinaryData.fromBytes(audioData)).subscribe();

4. Handle Events

session.receiveEvents().subscribe(event -> {
    ServerEventType eventType = event.getType();
    
    if (ServerEventType.SESSION_CREATED.equals(eventType)) {
        System.out.println("Session created");
    } else if (ServerEventType.INPUT_AUDIO_BUFFER_SPEECH_STARTED.equals(eventType)) {
        System.out.println("User started speaking");
    } else if (ServerEventType.INPUT_AUDIO_BUFFER_SPEECH_STOPPED.equals(eventType)) {
        System.out.println("User stopped speaking");
    } else if (ServerEventType.RESPONSE_AUDIO_DELTA.equals(eventType)) {
        if (event instanceof SessionUpdateResponseAudioDelta) {
            SessionUpdateResponseAudioDelta audioEvent = (SessionUpdateResponseAudioDelta) event;
            playAudioChunk(audioEvent.getDelta());
        }
    } else if (ServerEventType.RESPONSE_DONE.equals(eventType)) {
        System.out.println("Response complete");
    } else if (ServerEventType.ERROR.equals(eventType)) {
        if (event instanceof SessionUpdateError) {
            SessionUpdateError errorEvent = (SessionUpdateError) event;
            System.err.println("Error: " + errorEvent.getError().getMessage());
        }
    }
});

Voice Configuration

OpenAI Voices

// Available: ALLOY, ASH, BALLAD, CORAL, ECHO, SAGE, SHIMMER, VERSE
VoiceLiveSessionOptions options = new VoiceLiveSessionOptions()
    .setVoice(BinaryData.fromObject(new OpenAIVoice(OpenAIVoiceName.ALLOY)));

Azure Voices

// Azure Standard Voice
options.setVoice(BinaryData.fromObject(new AzureStandardVoice("en-US-JennyNeural")));

// Azure Custom Voice
options.setVoice(BinaryData.fromObject(new AzureCustomVoice("myVoice", "endpointId")));

// Azure Personal Voice
options.setVoice(BinaryData.fromObject(
    new AzurePersonalVoice("speakerProfileId", PersonalVoiceModels.PHOENIX_LATEST_NEURAL)));

Function Calling

VoiceLiveFunctionDefinition weatherFunction = new VoiceLiveFunctionDefinition("get_weather")
    .setDescription("Get current weather for a location")
    .setParameters(BinaryData.fromObject(parametersSchema));

VoiceLiveSessionOptions options = new VoiceLiveSessionOptions()
    .setTools(Arrays.asList(weatherFunction))
    .setInstructions("You have access to weather information.");

Best Practices

1. Use async client — VoiceLive requires reactive patterns

2. Configure turn detection for natural conversation flow

3. Enable noise reduction for better speech recognition

4. Handle interruptions gracefully with setInterruptResponse(true)

5. Use Whisper transcription for input audio transcription

6. Close sessions properly when conversation ends

Error Handling

session.receiveEvents()
    .doOnError(error -> System.err.println("Connection error: " + error.getMessage()))
    .onErrorResume(error -> {
        // Attempt reconnection or cleanup
        return Flux.empty();
    })
    .subscribe();

Reference Links

| Resource | URL |

|----------|-----|

| GitHub Source | https://github.com/Azure/azure-sdk-for-java/tree/main/sdk/ai/azure-ai-voicelive |

| Samples | https://github.com/Azure/azure-sdk-for-java/tree/main/sdk/ai/azure-ai-voicelive/src/samples |

When to Use

This skill is applicable to execute the workflow or actions described in the overview.

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

These are beta-version fragments, not a complete audio application. Verify endpoint/model/voice availability and SDK signatures against the deployed resource, supply imports/audio helpers and retain one managed event subscription with shutdown/error handling. Audio/transcription is transmitted to a cloud service: get consent for capture and transmission, minimize retention and protect transcripts. Function-call arguments are untrusted; validate them and authorize any external action separately before invoking it.