@mastra/voice-google 0.14.0 → 0.14.1-alpha.0

package/dist/docs/references/{docs-voice-overview.md → guides-voice-overview.md}

@@ -18,7 +18,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new OpenAIVoice(),
 })
 ```
@@ -29,7 +29,7 @@ You can then use the following voice capabilities:
 
 Turn your agent's responses into natural-sounding speech using Mastra's TTS capabilities. Choose from multiple providers like OpenAI, ElevenLabs, and more.
 
-For detailed configuration options and advanced features, check out our [Text-to-Speech guide](https://mastra.ai/docs/voice/text-to-speech).
+For detailed configuration options and advanced features, check out our [Text-to-Speech guide](https://mastra.ai/guides/voice/text-to-speech).
 
 **OpenAI**:
 
@@ -42,7 +42,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new OpenAIVoice(),
 })
 
@@ -70,7 +70,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new AzureVoice(),
 })
 
@@ -97,7 +97,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new ElevenLabsVoice(),
 })
 
@@ -124,7 +124,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new PlayAIVoice(),
 })
 
@@ -151,7 +151,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new GoogleVoice(),
 })
 
@@ -178,7 +178,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new CloudflareVoice(),
 })
 
@@ -205,7 +205,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new DeepgramVoice(),
 })
 
@@ -232,7 +232,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new InworldVoice(),
 })
 
@@ -259,7 +259,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new SpeechifyVoice(),
 })
 
@@ -286,7 +286,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new SarvamVoice(),
 })
 
@@ -313,7 +313,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new MurfVoice(),
 })
 
@@ -331,7 +331,7 @@ Visit the [Murf Voice Reference](https://mastra.ai/reference/voice/murf) for mor
 
 ### Speech to Text (STT)
 
-Transcribe spoken content using various providers like OpenAI, ElevenLabs, and more. For detailed configuration options and more, check out [Speech to Text](https://mastra.ai/docs/voice/speech-to-text).
+Transcribe spoken content using providers like OpenAI, ElevenLabs, and more. For detailed configuration options and more, check out [Speech to Text](https://mastra.ai/guides/voice/speech-to-text).
 
 You can download a sample audio file from [here](https://github.com/mastra-ai/realtime-voice-demo/raw/refs/heads/main/how_can_i_help_you.mp3).
 
@@ -348,7 +348,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new OpenAIVoice(),
 })
 
@@ -377,7 +377,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new AzureVoice(),
 })
 
@@ -405,7 +405,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new ElevenLabsVoice(),
 })
 
@@ -433,7 +433,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new GoogleVoice(),
 })
 
@@ -461,7 +461,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new CloudflareVoice(),
 })
 
@@ -489,7 +489,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new DeepgramVoice(),
 })
 
@@ -517,7 +517,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new InworldVoice(),
 })
 
@@ -545,7 +545,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new SarvamVoice(),
 })
 
@@ -564,7 +564,7 @@ Visit the [Sarvam Voice Reference](https://mastra.ai/reference/voice/sarvam) for
 
 ### Speech to Speech (STS)
 
-Create conversational experiences with speech-to-speech capabilities. The unified API enables real-time voice interactions between users and AI agents. For detailed configuration options and advanced features, check out [Speech to Speech](https://mastra.ai/docs/voice/speech-to-speech).
+Create conversational experiences with speech-to-speech capabilities. The unified API enables real-time voice interactions between users and AI agents. For detailed configuration options and advanced features, check out [Speech to Speech](https://mastra.ai/guides/voice/speech-to-speech).
 
 **OpenAI**:
 
@@ -577,7 +577,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new OpenAIRealtimeVoice(),
 })
 
@@ -607,7 +607,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new GeminiLiveVoice({
     // Live API mode
     apiKey: process.env.GOOGLE_API_KEY,
@@ -656,7 +656,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new NovaSonicVoice({
     region: 'us-east-1',
     speaker: 'matthew',
@@ -699,7 +699,7 @@ const voiceAgent = new Agent({
   id: 'voice-agent',
   name: 'Voice Agent',
   instructions: 'You are a voice assistant that can help users with their tasks.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice: new InworldRealtimeVoice({
     apiKey: process.env.INWORLD_API_KEY,
     model: 'inworld/models/gemma-4-26b-a4b-it',
@@ -773,6 +773,10 @@ await voiceAgent.voice.send(micStream)
 
 Visit the [xAI Realtime Voice Reference](https://mastra.ai/reference/voice/xai-realtime) for more information on the xAI voice provider.
 
+### Realtime voice
+
+Run live calls a user can talk over, in the browser or over the phone. Mastra hands the audio loop to LiveKit, which covers voice activity detection, semantic turn detection, and barge-in, while your agent generates each reply with its own model, tools, and memory. For setup and configuration options, check out [Realtime voice](https://mastra.ai/guides/voice/realtime-voice).
+
 ## Voice configuration
 
 Each voice provider can be configured with different models and options. Below are the detailed configuration options for all supported providers:
@@ -1134,7 +1138,7 @@ const voiceAgent = new Agent({
   id: 'aisdk-voice-agent',
   name: 'AI SDK Voice Agent',
   instructions: 'You are a helpful assistant with voice capabilities.',
-  model: 'openai/gpt-5.5',
+  model: 'openai/gpt-5.6-sol',
   voice,
 })
 ```
@@ -1248,5 +1252,6 @@ For more information on the CompositeVoice, refer to the [CompositeVoice Referen
 - [AWS Nova Sonic Voice](https://mastra.ai/reference/voice/aws-nova-sonic)
 - [Deepgram Voice](https://mastra.ai/reference/voice/deepgram)
 - [Inworld Voice](https://mastra.ai/reference/voice/inworld)
+- [LiveKit](https://mastra.ai/reference/voice/livekit)
 - [PlayAI Voice](https://mastra.ai/reference/voice/playai)
 - [Voice Examples](https://github.com/mastra-ai/voice-examples)

package/dist/docs/references/reference-voice-google.md

@@ -109,7 +109,17 @@ Converts speech to text using Google Cloud Speech-to-Text service. Supports both
 
 Pass `v2: true` to use the Cloud Speech-to-Text v2 API, which supports additional audio formats like AAC-in-MP4 (iOS Safari).
 
+The v2 `recognize` call is IAM-authorized and does not accept API-key-only authentication. Configure service account credentials on the `listeningModel` (or set `GOOGLE_APPLICATION_CREDENTIALS`) and set `GOOGLE_CLOUD_PROJECT` so the recognizer path can be resolved, even when `vertexAI` is not enabled.
+
 ```typescript
+import { GoogleVoice } from '@mastra/voice-google'
+
+// v2 listen() requires service account credentials, not just GOOGLE_API_KEY.
+// Set GOOGLE_CLOUD_PROJECT so the recognizer path can be resolved.
+const voice = new GoogleVoice({
+  listeningModel: { keyFilename: process.env.GOOGLE_APPLICATION_CREDENTIALS },
+})
+
 const transcript = await voice.listen(iosSafariAacStream, {
   v2: true,
   config: {
@@ -118,6 +128,8 @@ const transcript = await voice.listen(iosSafariAacStream, {
 })
 ```
 
+> **Note:** `listen({ v2: true })` fails with `PERMISSION_DENIED` on `speech.recognizers.recognize` when only `GOOGLE_API_KEY` is set. An API-key request carries no OAuth identity, so granting `roles/speech.client` to a user account does not help — the role must be granted to the service account presented in the request. This applies regardless of the `vertexAI` setting; `speak()` and v1 `listen()` still work with an API key alone.
+
 **audioStream** (`NodeJS.ReadableStream`): Audio stream to transcribe
 
 **options** (`GoogleListenOptionsV2`): v2 recognition options
@@ -162,7 +174,7 @@ The Google Voice provider supports two authentication methods:
 
 ### Standard Mode (API Key)
 
-Uses a Google Cloud API key for authentication. Suitable for development and basic use cases.
+Uses a Google Cloud API key for authentication. Covers `speak()` and v1 `listen()`. It does not cover `listen({ v2: true })`, which is IAM-authorized and requires service account credentials (see [v2](#v2)).
 
 ```typescript
 // Using environment variable (GOOGLE_API_KEY)
@@ -238,6 +250,8 @@ For Speech-to-Text:
 
 - `roles/speech.client` - Speech-to-Text Client
 
+Grant `roles/speech.client` to the service account whose credentials the request presents (via `keyFilename`, `credentials`, or `GOOGLE_APPLICATION_CREDENTIALS`). This role is required for `listen({ v2: true })` specifically, not only for Vertex AI mode. Granting it to a user account has no effect on API-key-only requests, which carry no identity to authorize.
+
 #### OAuth Scopes
 
 For synchronous Text-to-Speech synthesis:
@@ -269,6 +283,8 @@ For long-audio Text-to-Speech operations:
 
 6. The `listen()` method supports various recognition configurations through the Google Cloud Speech-to-Text API.
 
-7. Available voices can be filtered by language code using the `getSpeakers()` method.
+7. `listen({ v2: true })` requires service account credentials and `GOOGLE_CLOUD_PROJECT`; it fails with `PERMISSION_DENIED` when only `GOOGLE_API_KEY` is set. `speak()` and v1 `listen()` work with an API key alone.
+
+8. Available voices can be filtered by language code using the `getSpeakers()` method.
 
-8. Vertex AI mode provides enterprise features including IAM control, audit logs, and project-level billing.
+9. Vertex AI mode provides enterprise features including IAM control, audit logs, and project-level billing.