旧版天琴语音形象设计成型于2017年的天琴语音助手1.0，一直用到了3.0，在这几年里，语音技术获得了长足的进步，UI设计也有了新的趋势。早期的语音助手从功能交互层面只需传达语音助手工作的几个基本状态：等待唤醒、唤醒倾听、加载、播报；从视觉层面对标的是语音助手在手机上的呈现，如Siri，一来是为了在新世代的车机屏幕上为用户创造操作手机的熟悉感，二来是为了表现科技的进步为用户带来的全新体验，故将语音形象设计成一颗抽象光效风格的精灵球。
这一设计在当时的背景下是略微超前的。也正因为它的前瞻性，保证了它的“服役期”被大大延长。如今，语音技术早从之前的“语音识别+语义理解”进化出更丰富的认知能力，如声纹识别、情绪识别、性别识别、年龄识别、多音区等，辅以多模态交互能力，虚拟语音助手愈来愈趋近于真人的表现，它的身份也从语音助手逐步变成AI助理。在这一过程里，原先的设计就显得不够用了。
这是推出全新的天琴4.0语音形象的大致背景。
The old version of Lyra’s voice assistant avatar was formed in 2017 with the Lyra Voice Assistant 1.0 and has been used until version 3.0. Over these years, voice technology has made great progress and UI design has also followed new trends. Early voice assistants only needed to convey a few basic states of voice assistant operation from a functional interaction level: waiting for awakening, awakening listening, loading, and speaking; from a visual level, the benchmark was the presentation of the voice assistant on the phone, such as Siri. On the one hand, it was to create a familiar feeling for users when operating their phones on new-generation car screens; on the other hand, it was to show how technological progress brings new experiences to users. Therefore, the voice assistant avatar was designed as an abstract light effect style elf ball.
This design was slightly ahead of its time in its context. It is precisely because of its foresight that its “service life” has been greatly extended. Today, voice technology has evolved from “voice recognition + semantic understanding” to richer cognitive abilities such as voiceprint recognition, emotion recognition, gender recognition, age recognition and multi-tone zones etc., supplemented by multi-modal interaction capabilities. Virtual voice assistants are becoming closer and closer to human performance and their identity is gradually changing from a voice assistant to an AI assistant. In this process, the original design is no longer sufficient.
This is roughly the background for launching the brand new Lyra 4.0 Voice Assistant Avatar.
