Add Gandr as a provider to the Text-to-Speech block

Author: AALG123

apps/sim/blocks/blocks/tts.ts

@@ -11,7 +11,7 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
   description: 'Convert text to speech using AI voices',
   authMode: AuthMode.ApiKey,
   longDescription:
-    'Generate natural-sounding speech from text using state-of-the-art AI voices from OpenAI, Deepgram, ElevenLabs, Cartesia, Google Cloud, Azure, and PlayHT. Supports multiple voices, languages, and audio formats.',
+    'Generate natural-sounding speech from text using state-of-the-art AI voices from OpenAI, Deepgram, ElevenLabs, Cartesia, Google Cloud, Azure, PlayHT, and Gandr. Supports multiple voices, languages, and audio formats.',
   docsLink: 'https://docs.sim.ai/integrations/tts',
   category: 'blocks',
   integrationType: IntegrationType.AI,
@@ -41,6 +41,7 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
         { label: 'Google Cloud TTS', id: 'google' },
         { label: 'Azure TTS', id: 'azure' },
         { label: 'PlayHT', id: 'playht' },
+        { label: 'Gandr', id: 'gandr' },
       ],
       value: () => 'openai',
       required: true,
@@ -458,6 +459,41 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
       required: false,
     },
 
+    // Gandr Voice Selection
+    {
+      id: 'voice',
+      title: 'Voice',
+      type: 'dropdown',
+      condition: { field: 'provider', value: 'gandr' },
+      options: [
+        { label: 'Mia', id: 'gandr-mia' },
+        { label: 'Ava', id: 'gandr-ava' },
+        { label: 'Jenny', id: 'gandr-jenny' },
+        { label: 'Dane', id: 'gandr-dane' },
+        { label: 'Leo', id: 'gandr-leo' },
+        { label: 'Lewis', id: 'gandr-lewis' },
+      ],
+      value: () => 'gandr-mia',
+      dependsOn: ['provider'],
+      required: false,
+    },
+
+    // Gandr Response Format
+    {
+      id: 'responseFormat',
+      title: 'Audio Format',
+      type: 'dropdown',
+      condition: { field: 'provider', value: 'gandr' },
+      options: [
+        { label: 'MP3', id: 'mp3' },
+        { label: 'WAV', id: 'wav' },
+        { label: 'PCM', id: 'pcm' },
+      ],
+      value: () => 'mp3',
+      dependsOn: ['provider'],
+      required: false,
+    },
+
     // API Key (common to all providers)
     {
       id: 'apiKey',
@@ -478,6 +514,7 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
       'tts_google',
       'tts_azure',
       'tts_playht',
+      'tts_gandr',
     ],
     config: {
       tool: (params) => {
@@ -497,6 +534,8 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
             return 'tts_azure'
           case 'playht':
             return 'tts_playht'
+          case 'gandr':
+            return 'tts_gandr'
           default:
             return 'tts_openai'
         }
@@ -576,6 +615,14 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
           }
         }
 
+        if (params.provider === 'gandr') {
+          return {
+            ...baseParams,
+            voice: params.voice,
+            responseFormat: params.responseFormat,
+          }
+        }
+
         return baseParams
       },
     },
@@ -584,7 +631,7 @@ export const TtsBlock: BlockConfig<TtsBlockResponse> = {
   inputs: {
     provider: {
       type: 'string',
-      description: 'TTS provider (openai, deepgram, elevenlabs, cartesia, google, azure, playht)',
+      description: 'TTS provider (openai, deepgram, elevenlabs, cartesia, google, azure, playht, gandr)',
     },
     text: { type: 'string', description: 'Text to convert to speech' },
     apiKey: { type: 'string', description: 'Provider API key' },

apps/sim/lib/internal/tool-operations/registry.server.ts

@@ -818,6 +818,7 @@ const TTS_TOOL_IDS = [
   'tts_cartesia',
   'tts_deepgram',
   'tts_elevenlabs',
+  'tts_gandr',
   'tts_google',
   'tts_openai',
   'tts_playht',

apps/sim/lib/internal/tts/client.ts

@@ -16,6 +16,7 @@ import type {
   CartesiaTtsParams,
   DeepgramTtsParams,
   ElevenLabsTtsUnifiedParams,
+  GandrTtsParams,
   GoogleTtsParams,
   OpenAiTtsParams,
   PlayHtTtsParams,
@@ -137,6 +138,44 @@ export async function synthesizeOpenAi(
   }
 }
 
+export async function synthesizeGandr(
+  input: GandrTtsParams,
+  signal?: AbortSignal
+): Promise<TtsAudioResult> {
+  assertTextWithinLimit(input.text)
+  if (input.text.length > 2000) {
+    throw new TtsOperationError('Gandr TTS accepts up to 2000 characters per request', 400)
+  }
+  const voice = input.voice || 'gandr-mia'
+  const format = input.responseFormat || 'mp3'
+  const response = await providerFetch(
+    'https://tts.gandr.ai/v1/audio/speech',
+    {
+      method: 'POST',
+      headers: {
+        Authorization: `Bearer ${input.apiKey}`,
+        'Content-Type': 'application/json',
+      },
+      body: JSON.stringify({
+        model: 'tts-1',
+        input: input.text,
+        voice,
+        response_format: format,
+      }),
+    },
+    signal
+  )
+  if (!response.ok) {
+    const error = await readTtsErrorJson(response, 'Gandr TTS error response', signal)
+    throw new Error(`Gandr TTS API error: ${getTtsErrorMessage(error, response.statusText)}`)
+  }
+  return {
+    audioBuffer: await readAudio(response, 'Gandr TTS audio response', signal),
+    format,
+    mimeType: getTtsMimeType(format),
+  }
+}
+
 export async function synthesizeDeepgram(
   input: DeepgramTtsParams,
   signal?: AbortSignal

apps/sim/lib/internal/tts/execute-tool.ts

@@ -12,6 +12,7 @@ import {
   executeCartesiaTts,
   executeDeepgramTts,
   executeElevenLabsTts,
+  executeGandrTts,
   executeGoogleTts,
   executeLegacyElevenLabsTts,
   executeOpenAiTts,
@@ -114,6 +115,13 @@ const schemas = {
     textGuidance: z.coerce.number().optional(),
     sampleRate: z.coerce.number().optional(),
   }),
+  tts_gandr: z.object({
+    ...auth,
+    voice: z
+      .enum(['gandr-mia', 'gandr-ava', 'gandr-jenny', 'gandr-dane', 'gandr-leo', 'gandr-lewis'])
+      .optional(),
+    responseFormat: z.enum(['mp3', 'wav', 'pcm']).optional(),
+  }),
 } as const
 
 type TtsToolId = keyof typeof schemas
@@ -208,5 +216,7 @@ export const executeTtsTool: InternalToolOperationHandler = async (request) => {
       return executeOperation(schemas.tts_azure, request, executeAzureTts)
     case 'tts_playht':
       return executeOperation(schemas.tts_playht, request, executePlayHtTts)
+    case 'tts_gandr':
+      return executeOperation(schemas.tts_gandr, request, executeGandrTts)
   }
 }

apps/sim/lib/internal/tts/operations.ts

@@ -8,6 +8,7 @@ import {
   synthesizeCartesia,
   synthesizeDeepgram,
   synthesizeElevenLabs,
+  synthesizeGandr,
   synthesizeGoogle,
   synthesizeLegacyElevenLabs,
   synthesizeOpenAi,
@@ -21,6 +22,7 @@ import type { ElevenLabsTtsParams } from '@/tools/elevenlabs/types'
 import type {
   DeepgramTtsParams,
   ElevenLabsTtsUnifiedParams,
+  GandrTtsParams,
   GoogleTtsParams,
   PlayHtTtsParams,
   TtsProvider,
@@ -222,3 +224,15 @@ export async function executePlayHtTts(
     context
   )
 }
+
+export async function executeGandrTts(
+  input: GandrTtsParams,
+  context: TtsOperationContext
+): Promise<TtsResponse> {
+  return storeUnifiedAudio(
+    'gandr',
+    input.text,
+    await synthesizeGandr(input, context.signal),
+    context
+  )
+}

apps/sim/tools/registry.ts

@@ -5277,6 +5277,7 @@ import {
   cartesiaTtsTool,
   deepgramTtsTool,
   elevenLabsTtsUnifiedTool,
+  gandrTtsTool,
   googleTtsTool,
   openaiTtsTool,
   playhtTtsTool,
@@ -9303,6 +9304,7 @@ export const tools: Record<string, ExecutableToolConfig> = {
   tts_google: googleTtsTool,
   tts_azure: azureTtsTool,
   tts_playht: playhtTtsTool,
+  tts_gandr: gandrTtsTool,
   video_runway: runwayVideoTool,
   video_veo: veoVideoTool,
   video_luma: lumaVideoTool,

apps/sim/tools/tts/gandr.ts

@@ -0,0 +1,86 @@
+import type { GandrTtsParams, TtsBlockResponse } from '@/tools/tts/types'
+import type { InternalToolConfig } from '@/tools/types'
+
+export const gandrTtsTool: InternalToolConfig<GandrTtsParams, TtsBlockResponse> = {
+  id: 'tts_gandr',
+  name: 'Gandr TTS',
+  description: 'Convert text to speech using Gandr voices',
+  version: '1.0.0',
+
+  params: {
+    text: {
+      type: 'string',
+      required: true,
+      visibility: 'user-or-llm',
+      description:
+        'The text content to convert to speech, up to 2000 characters per request (e.g., "Hello, welcome to our service!")',
+    },
+    apiKey: {
+      type: 'string',
+      required: true,
+      visibility: 'user-only',
+      description: 'Gandr API key',
+    },
+    voice: {
+      type: 'string',
+      required: false,
+      visibility: 'user-or-llm',
+      description:
+        'Gandr voice identifier (e.g., "gandr-mia", "gandr-ava", "gandr-jenny", "gandr-dane", "gandr-leo", "gandr-lewis")',
+    },
+    responseFormat: {
+      type: 'string',
+      required: false,
+      visibility: 'user-only',
+      description: 'Audio format (mp3, wav, pcm)',
+    },
+  },
+
+  operation: {
+    modelInput: {
+      mode: 'project',
+      select: (params) => ({ text: params.text }),
+    },
+    input: (params) => ({
+      text: params.text,
+      apiKey: params.apiKey,
+      voice: params.voice || 'gandr-mia',
+      responseFormat: params.responseFormat || 'mp3',
+    }),
+  },
+
+  transformResponse: async (response: Response) => {
+    const data = await response.json()
+
+    if (!response.ok || data.error) {
+      return {
+        success: false,
+        error: data.error || 'TTS generation failed',
+        output: {
+          audioUrl: '',
+        },
+      }
+    }
+
+    return {
+      success: true,
+      output: {
+        audioUrl: data.audioUrl,
+        audioFile: data.audioFile,
+        duration: data.duration,
+        characterCount: data.characterCount,
+        format: data.format,
+        provider: data.provider,
+      },
+    }
+  },
+
+  outputs: {
+    audioUrl: { type: 'string', description: 'URL to the generated audio file' },
+    audioFile: { type: 'file', description: 'Generated audio file object' },
+    duration: { type: 'number', description: 'Audio duration in seconds' },
+    characterCount: { type: 'number', description: 'Number of characters processed' },
+    format: { type: 'string', description: 'Audio format' },
+    provider: { type: 'string', description: 'TTS provider used' },
+  },
+}

apps/sim/tools/tts/index.ts

@@ -2,6 +2,7 @@ export { azureTtsTool } from './azure'
 export { cartesiaTtsTool } from './cartesia'
 export { deepgramTtsTool } from './deepgram'
 export { elevenLabsTtsUnifiedTool } from './elevenlabs'
+export { gandrTtsTool } from './gandr'
 export { googleTtsTool } from './google'
 export { openaiTtsTool } from './openai'
 export { playhtTtsTool } from './playht'

apps/sim/tools/tts/operations.test.ts

@@ -7,6 +7,7 @@ import { azureTtsTool } from '@/tools/tts/azure'
 import { cartesiaTtsTool } from '@/tools/tts/cartesia'
 import { deepgramTtsTool } from '@/tools/tts/deepgram'
 import { elevenLabsTtsUnifiedTool } from '@/tools/tts/elevenlabs'
+import { gandrTtsTool } from '@/tools/tts/gandr'
 import { googleTtsTool } from '@/tools/tts/google'
 import { openaiTtsTool } from '@/tools/tts/openai'
 import { playhtTtsTool } from '@/tools/tts/playht'
@@ -20,6 +21,7 @@ const TOOLS = [
   googleTtsTool,
   azureTtsTool,
   playhtTtsTool,
+  gandrTtsTool,
 ] as const
 
 describe('TTS operation declarations', () => {
@@ -44,6 +46,12 @@ describe('TTS operation declarations', () => {
       responseFormat: 'mp3',
       speed: 1,
     })
+    expect(gandrTtsTool.operation.input({ text: 'Hello', apiKey: 'key' })).toEqual({
+      text: 'Hello',
+      apiKey: 'key',
+      voice: 'gandr-mia',
+      responseFormat: 'mp3',
+    })
     expect(
       elevenLabsTtsTool.operation.input({ text: 'Hello', apiKey: 'key', voiceId: 'voice_1' })
     ).toEqual({

apps/sim/tools/tts/types.ts

@@ -9,6 +9,7 @@ export type TtsProvider =
   | 'google'
   | 'azure'
   | 'playht'
+  | 'gandr'
 
 // OpenAI TTS Types
 export interface OpenAiTtsParams {
@@ -54,6 +55,14 @@ export interface ElevenLabsTtsUnifiedParams {
   apiKey: string
 }
 
+// Gandr TTS Types
+export interface GandrTtsParams {
+  text: string // up to 2000 characters per request
+  voice?: 'gandr-mia' | 'gandr-ava' | 'gandr-jenny' | 'gandr-dane' | 'gandr-leo' | 'gandr-lewis'
+  responseFormat?: 'mp3' | 'wav' | 'pcm'
+  apiKey: string
+}
+
 // Cartesia TTS Types
 export interface CartesiaTtsParams {
   text: string