程序包 io.agentscope.core.model.tts

类 DashScopeTTSModel

java.lang.Object

io.agentscope.core.model.tts.DashScopeTTSModel

所有已实现的接口:

TTSModel

public class DashScopeTTSModel
extends Object
implements TTSModel

DashScope TTS Model implementation.

This implementation uses direct HTTP calls to DashScope's TTS API, supporting multiple TTS models including:

• qwen3-tts-flash - Qwen3 TTS fast model
• qwen-tts - Qwen TTS model
• sambert-* - Sambert series (e.g., sambert-zhimao-v1)
• cosyvoice-* - CosyVoice series

Example usage:

 DashScopeTTSModel tts = DashScopeTTSModel.builder()
     .apiKey(System.getenv("DASHSCOPE_API_KEY"))
     .modelName("qwen3-tts-flash")
     .voice("Cherry")
     .build();

 TTSResponse response = tts.synthesize("你好，世界！", null).block();
 byte[] audioData = response.getAudioData();
 

另请参阅:

• DashScope TTS API Documentation

嵌套类概要

嵌套类

修饰符和类型	类	说明
static class	DashScopeTTSModel.Builder	Builder for constructing DashScopeTTSModel instances.

字段概要

字段

修饰符和类型	字段	说明
static final String	DEFAULT_BASE_URL	Default base URL for DashScope API.
static final String	TTS_ENDPOINT	TTS API endpoint (uses multimodal-generation for qwen-tts models).

方法概要

修饰符和类型	方法	说明
static DashScopeTTSModel.Builder	builder()	Creates a new builder for DashScopeTTSModel.
String	getModelName()	Returns the name of the TTS model being used.
reactor.core.publisher.Mono<TTSResponse>	synthesize(String text, TTSOptions options)	Synthesizes speech from text using DashScope TTS API.

从类继承的方法 java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

字段详细资料

DEFAULT_BASE_URL

public static final String DEFAULT_BASE_URL

Default base URL for DashScope API.

另请参阅:

• 常量字段值

TTS_ENDPOINT

public static final String TTS_ENDPOINT

TTS API endpoint (uses multimodal-generation for qwen-tts models).

另请参阅:

• 常量字段值

方法详细资料

synthesize

public reactor.core.publisher.Mono<TTSResponse> synthesize(String text,
 TTSOptions options)

Synthesizes speech from text using DashScope TTS API.

指定者:

synthesize 在接口中 TTSModel

参数:

text - the text to convert to speech

options - optional TTS configuration, uses defaults if null

返回:

a Mono containing the TTS response with audio data

getModelName

public String getModelName()

Returns the name of the TTS model being used.

指定者:

getModelName 在接口中 TTSModel

返回:

the model name (e.g., "qwen3-tts-flash", "qwen-tts")

builder

public static DashScopeTTSModel.Builder builder()

Creates a new builder for DashScopeTTSModel.

返回:

a new Builder instance