Files
moeka-project/components/MainStage.vue
T

378 lines
12 KiB
Vue

<script setup lang="ts">
import type {
CoreAssistantMessage,
CoreSystemMessage,
CoreUserMessage,
} from 'ai'
import type {
Emotion,
} from '../constants/emotions'
import { useLocalStorage } from '@vueuse/core'
import { computed, onMounted, ref, watch } from 'vue'
import Avatar from '../assets/live2d/models/hiyori_free_zh/avatar.png'
import { useMarkdown } from '../composables/markdown'
import { useQueue } from '../composables/queue'
import {
useDelayMessageQueue,
useEmotionsMessageQueue,
useMessageContentQueue,
} from '../composables/queues'
import { llmInferenceEndToken } from '../constants'
import {
EMOTION_EmotioMotionName_value,
EmotionThinkMotionName,
} from '../constants/emotions'
import SystemPromptV2 from '../constants/prompts/system-v2'
import { useLLM } from '../stores/llm'
import BasicTextarea from './BasicTextarea.vue'
// import AudioWaveform from './AudioWaveform.vue'
import Live2DViewer from './Live2DViewer.vue'
const nowSpeakingAvatarBorderOpacityMin = 30
const nowSpeakingAvatarBorderOpacityMax = 100
const openAiApiKey = useLocalStorage('openai-api-key', '')
const openAiApiBaseURL = useLocalStorage('openai-api-base-url', '')
const openAIModel = useLocalStorage<{ id: string, name?: string }>('openai-model', { id: 'openai/gpt-3.5-turbo', name: 'OpenAI GPT3.5 Turbo' })
const { setupOpenAI, streamSpeech, stream, models } = useLLM()
const { audioContext, calculateVolume } = useAudioContext()
const { process } = useMarkdown()
const listening = ref(false)
const live2DViewerRef = ref<{ setMotion: (motionName: string) => Promise<void> }>()
const supportedModels = ref<{ id: string, name?: string }[]>([])
const messageInput = ref<string>('')
const messages = ref<Array<CoreAssistantMessage | CoreUserMessage | CoreSystemMessage>>([SystemPromptV2 as CoreSystemMessage])
const streamingMessage = ref<CoreAssistantMessage>({ role: 'assistant', content: '' })
const audioAnalyser = ref<AnalyserNode>()
const mouthOpenSize = ref(0)
const nowSpeaking = ref(false)
const model = ref('')
const lipSyncStarted = ref(false)
const nowSpeakingAvatarBorderOpacity = computed<number>(() => {
if (!nowSpeaking.value)
return nowSpeakingAvatarBorderOpacityMin
return ((nowSpeakingAvatarBorderOpacityMin
+ (nowSpeakingAvatarBorderOpacityMax - nowSpeakingAvatarBorderOpacityMin) * mouthOpenSize.value) / 100)
})
function handleModelChange(event: Event) {
const target = event.target as HTMLSelectElement
const found = supportedModels.value.find(m => m.id === target.value)
if (!found) {
openAIModel.value = undefined
return
}
openAIModel.value = found
}
const audioQueue = useQueue<{ audioBuffer: AudioBuffer, text: string }>({
handlers: [
(ctx) => {
return new Promise((resolve) => {
// Create an AudioBufferSourceNode
const source = audioContext.createBufferSource()
source.buffer = ctx.data.audioBuffer
// Connect the source to the AudioContext's destination (the speakers)
source.connect(audioContext.destination)
// Connect the source to the analyzer
source.connect(audioAnalyser.value!)
// Start playing the audio
nowSpeaking.value = true
source.start(0)
source.onended = () => {
nowSpeaking.value = false
resolve()
}
})
},
],
})
const ttsQueue = useQueue<string>({
handlers: [
async (ctx) => {
const now = Date.now()
const res = await streamSpeech(ctx.data)
const elapsed = Date.now() - now
// eslint-disable-next-line no-console
console.debug('TTS took', elapsed, 'ms')
// Decode the ArrayBuffer into an AudioBuffer
const audioBuffer = await audioContext.decodeAudioData(res)
await audioQueue.add({ audioBuffer, text: ctx.data })
},
],
})
ttsQueue.on('add', (content) => {
// eslint-disable-next-line no-console
console.debug('ttsQueue added', content)
})
const messageContentQueue = useMessageContentQueue(ttsQueue)
const emotionsQueue = useQueue<Emotion>({
handlers: [
async (ctx) => {
await live2DViewerRef.value!.setMotion(EMOTION_EmotioMotionName_value[ctx.data])
},
],
})
const emotionMessageContentQueue = useEmotionsMessageQueue(emotionsQueue, messageContentQueue)
emotionMessageContentQueue.onHandlerEvent('emotion', (emotion) => {
// eslint-disable-next-line no-console
console.debug('emotion detected', emotion)
})
const delaysQueue = useDelayMessageQueue(emotionMessageContentQueue)
delaysQueue.onHandlerEvent('delay', (delay) => {
// eslint-disable-next-line no-console
console.debug('delay detected', delay)
})
function getVolumeWithMinMaxNormalizeWithFrameUpdates() {
requestAnimationFrame(getVolumeWithMinMaxNormalizeWithFrameUpdates)
if (!nowSpeaking.value)
return
mouthOpenSize.value = calculateVolume(audioAnalyser.value!, 'linear')
}
function setupLipSync() {
if (!lipSyncStarted.value) {
getVolumeWithMinMaxNormalizeWithFrameUpdates()
audioContext.resume()
lipSyncStarted.value = true
}
}
function setupAnalyser() {
if (!audioAnalyser.value)
audioAnalyser.value = audioContext.createAnalyser()
}
async function onSendMessage(sendingMessage: string) {
if (!sendingMessage)
return
setupLipSync()
setupAnalyser()
streamingMessage.value = { role: 'assistant', content: '' }
messages.value.push({ role: 'user', content: sendingMessage })
messages.value.push(streamingMessage.value)
// const index = messages.value.length - 1
live2DViewerRef.value?.setMotion(EmotionThinkMotionName)
const res = await stream(model.value, messages.value.slice(0, messages.value.length - 1))
enum States {
Literal = 'literal',
Special = 'special',
}
let state = States.Literal
let buffer = ''
for await (const textPart of res.textStream) {
for (const textSingleChar of textPart) {
let newState: States = state
if (textSingleChar === '<')
newState = States.Special
else if (textSingleChar === '>')
newState = States.Literal
if (state === States.Literal && newState === States.Special) {
streamingMessage.value.content += buffer
buffer = ''
}
if (state === States.Special && newState === States.Literal)
buffer = '' // Clear buffer when exiting Special state
if (state === States.Literal && newState === States.Literal) {
streamingMessage.value.content += textSingleChar
buffer = ''
}
await delaysQueue.add(textSingleChar)
state = newState
buffer += textSingleChar
}
}
if (buffer)
streamingMessage.value.content += buffer
await delaysQueue.add(llmInferenceEndToken)
messageInput.value = ''
}
watch(openAiApiKey, async (value) => {
setupOpenAI({
apiKey: value,
baseURL: openAiApiBaseURL.value,
})
const fetchedModels = await models()
supportedModels.value = fetchedModels.data
})
onMounted(async () => {
if (!openAiApiKey.value)
return
setupOpenAI({
apiKey: openAiApiKey.value,
baseURL: openAiApiBaseURL.value,
})
const fetchedModels = await models()
supportedModels.value = fetchedModels.data
})
onUnmounted(() => {
lipSyncStarted.value = false
})
</script>
<template>
<div max-h="[100vh]" h-full p="2" flex="~ col">
<div space-x="2" flex="~ row" w-full>
<div flex="~ row" w-full>
<input
v-model="openAiApiKey"
placeholder="Input your API key"
p="2" bg="zinc-100 dark:zinc-700" w-full rounded-lg outline-none
>
</div>
<div flex="~ row" w-full>
<input
v-model="openAiApiBaseURL"
placeholder="Input your API base URL"
p="2" bg="zinc-100 dark:zinc-700" w-full rounded-lg outline-none
>
</div>
</div>
<div flex="~ row 1" w-full items-end space-x-2>
<div w-full min-h="100 sm:100">
<Live2DViewer ref="live2DViewerRef" :mouth-open-size="mouthOpenSize" model="assets/live2d/models/hiyori_pro_zh/runtime/hiyori_pro_t11.model3.json" />
<!-- <div>
<input v-model.number="mouthOpenSize" type="range" max="1" min="0" step="0.01">
<span>{{ mouthOpenSize }}</span>
</div> -->
<!-- <AudioWaveform ref="audioWaveformRef" /> -->
</div>
<div my="2" w-full space-y-2 max-h="[calc(100vh-117px)]">
<div v-for="(message, index) in messages" :key="index">
<div v-if="message.role === 'assistant'" flex mr="12">
<div
mr-2 h-10 min-h-10 min-w-10 w-10 overflow-hidden rounded-full
border="solid 3"
transition="all ease-in-out" duration-100
:style="{
borderColor: `rgba(236, 72, 153, ${nowSpeakingAvatarBorderOpacity.toFixed(2)})`,
}"
>
<img :src="Avatar">
</div>
<div flex="~ col" bg="pink-50/50 dark:pink-900/50" p="2" border="2 solid pink/10" rounded-lg>
<div>
<span font-semibold>Neuro</span>
</div>
<div v-html="process(message.content as string)" />
</div>
</div>
<div v-else-if="message.role === 'user'" flex="~ row-reverse" ml="12">
<div border="purple solid 3" ml="2" h-10 min-h-10 min-w-10 w-10 overflow-hidden rounded-full>
<div i-carbon:user-avatar-filled text="purple" h-full w-full p="0" m="0" />
</div>
<div flex="~ col" bg="purple-50/50 dark:purple-900/50" p="2" border="2 solid pink/10" rounded-lg>
<div>
<span font-semibold>You</span>
</div>
<div v-html="process(message.content as string)" />
</div>
</div>
</div>
</div>
</div>
<div my="2" space-x="2" flex="~ row" w-full self-end>
<div flex="~ col" w-full space-y="2">
<select
p="2"
bg="zinc-100 dark:zinc-700" w-full rounded-lg
outline-none
@change="handleModelChange"
>
<option disabled>
Select a model
</option>
<option v-if="openAIModel" :value="openAIModel.id">
{{ 'name' in openAIModel ? `${openAIModel.name} (${openAIModel.id})` : openAIModel.id }}
</option>
<option v-for="m in supportedModels" :key="m.id" :value="m.id">
{{ 'name' in m ? `${m.name} (${m.id})` : m.id }}
</option>
</select>
<div absolute bottom="5" left="50%" translate-x="-50%">
<button
bg="zinc-100 dark:zinc-700" flex="~ row"
items-center rounded-full px-4 py-2
transition="all ease-in-out"
@click="listening = !listening"
>
<Transition mode="out-in">
<div v-if="listening" flex="~ row" items-center space-x-1>
<div i-carbon:microphone-filled text-red />
<span>
Listening...
</span>
</div>
<div v-else flex="~ row" items-center space-x-1>
<div i-carbon:microphone text-inherit />
<span>
Talk
</span>
</div>
</Transition>
</button>
</div>
<BasicTextarea
v-model="messageInput"
placeholder="Message"
p="2" bg="zinc-100 dark:zinc-700"
w-full rounded-lg outline-none
@submit="onSendMessage"
/>
</div>
</div>
</div>
</template>
<style>
.v-enter-active,
.v-leave-active {
transition: opacity 0.5s ease;
}
.v-enter-from,
.v-leave-to {
opacity: 0;
}
</style>