【尝鲜】SpringCloudAlibaba AI 配置&使用教程

1、环境配置

• maven依赖pom.xml
  注意配置远程仓库，原因见：Unresolved dependency: ‘org.springframework.ai:spring-ai-core:jar:0.8.1’

<dependencies>

		<!--Base-->
		<dependency>
			<groupId>org.springframework.boot</groupId>
			<artifactId>spring-boot-starter-web</artifactId>
		</dependency>
		<dependency>
			<groupId>org.springframework.boot</groupId>
			<artifactId>spring-boot-starter-test</artifactId>
			<scope>test</scope>
		</dependency>

		<!--AI-->
		<dependency>
			<groupId>com.alibaba.cloud</groupId>
			<artifactId>spring-cloud-starter-alibaba-ai</artifactId>
			<version>2023.0.1.0</version>
			<exclusions>
                <exclusion>
                    <groupId>org.slf4j</groupId>
                    <artifactId>slf4j-simple</artifactId>
                </exclusion>
            </exclusions>
		</dependency>

	</dependencies>

	<repositories>
        <repository>
            <id>spring-milestones</id>
            <name>Spring Milestones</name>
            <url>https://repo.spring.io/milestone</url>
            <snapshots>
                <enabled>false</enabled>
            </snapshots>
        </repository>
    </repositories>

• application.yaml

# Tomcat
server:
  port: 9310

# Spring
spring:
  cloud:
    ai:
      tongyi:
          api-key: api密钥

api-key申请地址：申请地址

2、服务实现

controller就不详细展示了，主要展示service中如何调用

@Service
@AllArgsConstructor
public class TongyiServiceImpl implements ITongyiService {

    private final ChatClient chatClient;
    private final ImageClient imageClient;
    private final SpeechClient speechClient;


	// 对话
    @Override
    public String chat(String question) {
        return chatClient.call(question);
    }
	
	// 文生图
    @Override
    public String image(String promptWord) {
        ImagePrompt imagePrompt = new ImagePrompt(promptWord);
        return imageClient.call(imagePrompt).getResult().getOutput().getUrl();
    }

	// 语音
    @Override
    public ByteBuffer speech(String promptWord) {
        SpeechPrompt question = new SpeechPrompt(promptWord);
        return speechClient.call(question).getResult().getOutput();
    }

}

效果图