chore: bump xsai, update speech params (#12)

moeru-ai · Jan 9, 2025 · e0bae0b · e0bae0b
1 parent acf0d63
commit e0bae0b
Show file tree

Hide file tree

Showing 7 changed files with 108 additions and 81 deletions.
diff --git a/packages/stage-tamagotchi/package.json b/packages/stage-tamagotchi/package.json
@@ -51,12 +51,12 @@
     "@vueuse/core": "^12.3.0",
     "@vueuse/head": "^2.0.0",
     "@vueuse/shared": "^12.3.0",
-    "@xsai/generate-speech": "^0.0.24",
-    "@xsai/generate-text": "^0.0.24",
-    "@xsai/model": "^0.0.24",
-    "@xsai/providers": "^0.0.24",
-    "@xsai/shared-chat": "^0.0.24",
-    "@xsai/stream-text": "^0.0.24",
+    "@xsai/generate-speech": "catalog:",
+    "@xsai/generate-text": "catalog:",
+    "@xsai/model": "catalog:",
+    "@xsai/providers": "catalog:",
+    "@xsai/shared-chat": "catalog:",
+    "@xsai/stream-text": "catalog:",
     "defu": "^6.1.4",
     "jszip": "^3.10.1",
     "nprogress": "^0.2.0",

diff --git a/packages/stage-ui/src/components/Widgets/Stage.vue b/packages/stage-ui/src/components/Widgets/Stage.vue
@@ -77,15 +77,14 @@ const ttsQueue = useQueue<string>({
         baseURL: 'https://unspeech.hyp3r.link/v1/',
       })
       const res = await generateSpeech({
-        ...elevenlabs.speech({
-          model: 'elevenlabs/eleven_multilingual_v2',
-          voice: voiceMap[voice],
+        ...elevenlabs.speech('eleven_multilingual_v2', {
           voiceSettings: {
             stability: 0.4,
             similarityBoost: 0.5,
           },
         }),
         input: ctx.data,
+        voice: voiceMap[voice],
       })
       const elapsed = Date.now() - now
 

diff --git a/packages/stage-web/package.json b/packages/stage-web/package.json
@@ -48,12 +48,12 @@
     "@vueuse/core": "^12.3.0",
     "@vueuse/head": "^2.0.0",
     "@vueuse/shared": "^12.3.0",
-    "@xsai/generate-speech": "^0.0.24",
-    "@xsai/generate-text": "^0.0.24",
-    "@xsai/model": "^0.0.24",
-    "@xsai/providers": "^0.0.24",
-    "@xsai/shared-chat": "^0.0.24",
-    "@xsai/stream-text": "^0.0.24",
+    "@xsai/generate-speech": "catalog:",
+    "@xsai/generate-text": "catalog:",
+    "@xsai/model": "catalog:",
+    "@xsai/providers": "catalog:",
+    "@xsai/shared-chat": "catalog:",
+    "@xsai/stream-text": "catalog:",
     "defu": "^6.1.4",
     "jszip": "^3.10.1",
     "nprogress": "^0.2.0",

diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml
diff --git a/pnpm-workspace.yaml b/pnpm-workspace.yaml
@@ -3,3 +3,11 @@ packages:
   - 'services/**'
   - 'examples/**'
   - '!**/dist/**'
+
+catalog:
+  '@xsai/generate-speech': ^0.0.26
+  '@xsai/generate-text': ^0.0.26
+  '@xsai/model': ^0.0.26
+  '@xsai/providers': ^0.0.26
+  '@xsai/shared-chat': ^0.0.26
+  '@xsai/stream-text': ^0.0.26
diff --git a/services/discord-voice-bot/package.json b/services/discord-voice-bot/package.json
@@ -36,10 +36,10 @@
     "@dotenvx/dotenvx": "^1.32.1",
     "@guiiai/logg": "^1.0.6",
     "@huggingface/transformers": "^3.2.4",
-    "@xsai/generate-speech": "^0.0.24",
-    "@xsai/generate-text": "^0.0.24",
-    "@xsai/providers": "^0.0.24",
-    "@xsai/shared-chat": "^0.0.24",
+    "@xsai/generate-speech": "catalog:",
+    "@xsai/generate-text": "catalog:",
+    "@xsai/providers": "catalog:",
+    "@xsai/shared-chat": "catalog:",
     "discord.js": "^14.17.3",
     "libsodium-wrappers": "^0.7.15",
     "opusscript": "^0.1.1",

diff --git a/services/discord-voice-bot/src/bots/discord/commands/summon.ts b/services/discord-voice-bot/src/bots/discord/commands/summon.ts
@@ -155,15 +155,14 @@ export async function handleSummon(log: ReturnType<typeof useLogg>, interaction:
         })
 
         const speechRes = await generateSpeech({
-          ...elevenlabs.speech({
-            model: 'eleven_multilingual_v2',
-            voice: 'lNxY9WuCBCZCISASyJ55',
+          ...elevenlabs.speech('eleven_multilingual_v2', {
             voiceSettings: {
               stability: 0.4,
               similarityBoost: 0.5,
             },
           }),
           input: res.text,
+          voice: 'lNxY9WuCBCZCISASyJ55',
         })
 
         log.withField('length', speechRes.byteLength).log('Generated speech')