@mastra/voice-murf 0.12.2-alpha.0 → 0.12.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +8 -0
- package/dist/_types/@internal_voice/dist/_types/@internal_ai-sdk-v5/dist/index.d.ts +4 -159
- package/dist/docs/SKILL.md +1 -1
- package/dist/docs/assets/SOURCE_MAP.json +1 -1
- package/dist/docs/references/docs-agents-adding-voice.md +37 -6
- package/dist/docs/references/docs-voice-overview.md +86 -24
- package/package.json +4 -4
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,13 @@
|
|
|
1
1
|
# @mastra/voice-murf
|
|
2
2
|
|
|
3
|
+
## 0.12.2
|
|
4
|
+
|
|
5
|
+
### Patch Changes
|
|
6
|
+
|
|
7
|
+
- Moved shared voice primitives and route metadata into the new `@internal/voice` package so voice providers no longer depend on `@mastra/core` and server voice routes share the same route definitions. ([#16725](https://github.com/mastra-ai/mastra/pull/16725))
|
|
8
|
+
|
|
9
|
+
`@mastra/core/voice` continues to re-export the voice APIs for backwards compatibility.
|
|
10
|
+
|
|
3
11
|
## 0.12.2-alpha.0
|
|
4
12
|
|
|
5
13
|
### Patch Changes
|
|
@@ -1,3 +1,5 @@
|
|
|
1
|
+
import { JSONSchema7 } from 'json-schema';
|
|
2
|
+
import { JSONSchema7Definition } from 'json-schema';
|
|
1
3
|
import { ServerResponse } from 'node:http';
|
|
2
4
|
import { ServerResponse as ServerResponse_2 } from 'http';
|
|
3
5
|
import { z } from '../../zod/v4/index.d.cts';
|
|
@@ -3507,164 +3509,7 @@ export declare function jsonSchema<OBJECT = unknown>(jsonSchema: JSONSchema7 | (
|
|
|
3507
3509
|
validate?: (value: unknown) => ValidationResult<OBJECT> | PromiseLike<ValidationResult<OBJECT>>;
|
|
3508
3510
|
}): Schema<OBJECT>;
|
|
3509
3511
|
|
|
3510
|
-
export
|
|
3511
|
-
$id?: string | undefined;
|
|
3512
|
-
$ref?: string | undefined;
|
|
3513
|
-
$schema?: JSONSchema7Version | undefined;
|
|
3514
|
-
$comment?: string | undefined;
|
|
3515
|
-
|
|
3516
|
-
/**
|
|
3517
|
-
* @see https://datatracker.ietf.org/doc/html/draft-bhutton-json-schema-00#section-8.2.4
|
|
3518
|
-
* @see https://datatracker.ietf.org/doc/html/draft-bhutton-json-schema-validation-00#appendix-A
|
|
3519
|
-
*/
|
|
3520
|
-
$defs?: {
|
|
3521
|
-
[key: string]: JSONSchema7Definition;
|
|
3522
|
-
} | undefined;
|
|
3523
|
-
|
|
3524
|
-
/**
|
|
3525
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.1
|
|
3526
|
-
*/
|
|
3527
|
-
type?: JSONSchema7TypeName | JSONSchema7TypeName[] | undefined;
|
|
3528
|
-
enum?: JSONSchema7Type[] | undefined;
|
|
3529
|
-
const?: JSONSchema7Type | undefined;
|
|
3530
|
-
|
|
3531
|
-
/**
|
|
3532
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.2
|
|
3533
|
-
*/
|
|
3534
|
-
multipleOf?: number | undefined;
|
|
3535
|
-
maximum?: number | undefined;
|
|
3536
|
-
exclusiveMaximum?: number | undefined;
|
|
3537
|
-
minimum?: number | undefined;
|
|
3538
|
-
exclusiveMinimum?: number | undefined;
|
|
3539
|
-
|
|
3540
|
-
/**
|
|
3541
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.3
|
|
3542
|
-
*/
|
|
3543
|
-
maxLength?: number | undefined;
|
|
3544
|
-
minLength?: number | undefined;
|
|
3545
|
-
pattern?: string | undefined;
|
|
3546
|
-
|
|
3547
|
-
/**
|
|
3548
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.4
|
|
3549
|
-
*/
|
|
3550
|
-
items?: JSONSchema7Definition | JSONSchema7Definition[] | undefined;
|
|
3551
|
-
additionalItems?: JSONSchema7Definition | undefined;
|
|
3552
|
-
maxItems?: number | undefined;
|
|
3553
|
-
minItems?: number | undefined;
|
|
3554
|
-
uniqueItems?: boolean | undefined;
|
|
3555
|
-
contains?: JSONSchema7Definition | undefined;
|
|
3556
|
-
|
|
3557
|
-
/**
|
|
3558
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.5
|
|
3559
|
-
*/
|
|
3560
|
-
maxProperties?: number | undefined;
|
|
3561
|
-
minProperties?: number | undefined;
|
|
3562
|
-
required?: string[] | undefined;
|
|
3563
|
-
properties?: {
|
|
3564
|
-
[key: string]: JSONSchema7Definition;
|
|
3565
|
-
} | undefined;
|
|
3566
|
-
patternProperties?: {
|
|
3567
|
-
[key: string]: JSONSchema7Definition;
|
|
3568
|
-
} | undefined;
|
|
3569
|
-
additionalProperties?: JSONSchema7Definition | undefined;
|
|
3570
|
-
dependencies?: {
|
|
3571
|
-
[key: string]: JSONSchema7Definition | string[];
|
|
3572
|
-
} | undefined;
|
|
3573
|
-
propertyNames?: JSONSchema7Definition | undefined;
|
|
3574
|
-
|
|
3575
|
-
/**
|
|
3576
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.6
|
|
3577
|
-
*/
|
|
3578
|
-
if?: JSONSchema7Definition | undefined;
|
|
3579
|
-
then?: JSONSchema7Definition | undefined;
|
|
3580
|
-
else?: JSONSchema7Definition | undefined;
|
|
3581
|
-
|
|
3582
|
-
/**
|
|
3583
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.7
|
|
3584
|
-
*/
|
|
3585
|
-
allOf?: JSONSchema7Definition[] | undefined;
|
|
3586
|
-
anyOf?: JSONSchema7Definition[] | undefined;
|
|
3587
|
-
oneOf?: JSONSchema7Definition[] | undefined;
|
|
3588
|
-
not?: JSONSchema7Definition | undefined;
|
|
3589
|
-
|
|
3590
|
-
/**
|
|
3591
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-7
|
|
3592
|
-
*/
|
|
3593
|
-
format?: string | undefined;
|
|
3594
|
-
|
|
3595
|
-
/**
|
|
3596
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-8
|
|
3597
|
-
*/
|
|
3598
|
-
contentMediaType?: string | undefined;
|
|
3599
|
-
contentEncoding?: string | undefined;
|
|
3600
|
-
|
|
3601
|
-
/**
|
|
3602
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-9
|
|
3603
|
-
*/
|
|
3604
|
-
definitions?: {
|
|
3605
|
-
[key: string]: JSONSchema7Definition;
|
|
3606
|
-
} | undefined;
|
|
3607
|
-
|
|
3608
|
-
/**
|
|
3609
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-10
|
|
3610
|
-
*/
|
|
3611
|
-
title?: string | undefined;
|
|
3612
|
-
description?: string | undefined;
|
|
3613
|
-
default?: JSONSchema7Type | undefined;
|
|
3614
|
-
readOnly?: boolean | undefined;
|
|
3615
|
-
writeOnly?: boolean | undefined;
|
|
3616
|
-
examples?: JSONSchema7Type | undefined;
|
|
3617
|
-
}
|
|
3618
|
-
|
|
3619
|
-
declare interface JSONSchema7Array extends Array<JSONSchema7Type> {}
|
|
3620
|
-
|
|
3621
|
-
/**
|
|
3622
|
-
* JSON Schema v7
|
|
3623
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01
|
|
3624
|
-
*/
|
|
3625
|
-
declare type JSONSchema7Definition = JSONSchema7 | boolean;
|
|
3626
|
-
|
|
3627
|
-
declare interface JSONSchema7Object {
|
|
3628
|
-
[key: string]: JSONSchema7Type;
|
|
3629
|
-
}
|
|
3630
|
-
|
|
3631
|
-
/**
|
|
3632
|
-
* Primitive type
|
|
3633
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.1.1
|
|
3634
|
-
*/
|
|
3635
|
-
declare type JSONSchema7Type =
|
|
3636
|
-
| string //
|
|
3637
|
-
| number
|
|
3638
|
-
| boolean
|
|
3639
|
-
| JSONSchema7Object
|
|
3640
|
-
| JSONSchema7Array
|
|
3641
|
-
| null;
|
|
3642
|
-
|
|
3643
|
-
/**
|
|
3644
|
-
* Primitive type
|
|
3645
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-6.1.1
|
|
3646
|
-
*/
|
|
3647
|
-
declare type JSONSchema7TypeName =
|
|
3648
|
-
| "string" //
|
|
3649
|
-
| "number"
|
|
3650
|
-
| "integer"
|
|
3651
|
-
| "boolean"
|
|
3652
|
-
| "object"
|
|
3653
|
-
| "array"
|
|
3654
|
-
| "null";
|
|
3655
|
-
|
|
3656
|
-
/**
|
|
3657
|
-
* Meta schema
|
|
3658
|
-
*
|
|
3659
|
-
* Recommended values:
|
|
3660
|
-
* - 'http://json-schema.org/schema#'
|
|
3661
|
-
* - 'http://json-schema.org/hyper-schema#'
|
|
3662
|
-
* - 'http://json-schema.org/draft-07/schema#'
|
|
3663
|
-
* - 'http://json-schema.org/draft-07/hyper-schema#'
|
|
3664
|
-
*
|
|
3665
|
-
* @see https://tools.ietf.org/html/draft-handrews-json-schema-validation-01#section-5
|
|
3666
|
-
*/
|
|
3667
|
-
declare type JSONSchema7Version = string;
|
|
3512
|
+
export { JSONSchema7 }
|
|
3668
3513
|
|
|
3669
3514
|
export declare class JsonToSseTransformStream extends TransformStream<unknown, string> {
|
|
3670
3515
|
constructor();
|
|
@@ -9039,5 +8884,5 @@ export declare function zodSchema<OBJECT>(zodSchema: z4.core.$ZodType<OBJECT, an
|
|
|
9039
8884
|
}): Schema<OBJECT>;
|
|
9040
8885
|
|
|
9041
8886
|
export { }
|
|
9042
|
-
export { GatewayProviderSettings as GatewayProviderSettings, GatewayProvider as GatewayProvider, ParseResult as ParseResult, FlexibleValidator as FlexibleValidator, ValidationResult as ValidationResult, LanguageModelV2ToolResultOutput as LanguageModelV2ToolResultOutput, ProviderOptions as ProviderOptions, ReasoningPart as ReasoningPart, ToolOutputProperties as ToolOutputProperties, LanguageModelV2ToolResultPart as LanguageModelV2ToolResultPart, FlexibleSchema as FlexibleSchema,
|
|
8887
|
+
export { GatewayProviderSettings as GatewayProviderSettings, GatewayProvider as GatewayProvider, ParseResult as ParseResult, FlexibleValidator as FlexibleValidator, ValidationResult as ValidationResult, LanguageModelV2ToolResultOutput as LanguageModelV2ToolResultOutput, ProviderOptions as ProviderOptions, ReasoningPart as ReasoningPart, ToolOutputProperties as ToolOutputProperties, LanguageModelV2ToolResultPart as LanguageModelV2ToolResultPart, FlexibleSchema as FlexibleSchema, ImageModelV2CallWarning as ImageModelV2CallWarning, SpeechModelV2CallWarning as SpeechModelV2CallWarning, TranscriptionModelV2CallWarning as TranscriptionModelV2CallWarning, AttributeValue as AttributeValue, Tracer as Tracer, EmbeddingModelV2Embedding as EmbeddingModelV2Embedding, ImageModelV2ProviderMetadata as ImageModelV2ProviderMetadata, GlobalProviderModelId as GlobalProviderModelId, LanguageModelV2FinishReason as LanguageModelV2FinishReason, LanguageModelV2CallWarning as LanguageModelV2CallWarning, LanguageModelV2Middleware as LanguageModelV2Middleware, SharedV2ProviderMetadata as SharedV2ProviderMetadata, LanguageModelV2Usage as LanguageModelV2Usage, Source as Source, ResponseMessage as ResponseMessage, DeepPartialInternal as DeepPartialInternal, LanguageModelV2ToolCall as LanguageModelV2ToolCall, StreamTextOnAbortCallback as StreamTextOnAbortCallback, ValueOf as ValueOf, asUITool as asUITool, _ai_sdk_provider_utils as _ai_sdk_provider_utils, _ai_sdk_provider as _ai_sdk_provider, DataUIMessageChunk as DataUIMessageChunk, ConsumeStreamOptions as ConsumeStreamOptions, Output_2 as Output_2, InferAgentTools as InferAgentTools, SingleRequestTextStreamPart as SingleRequestTextStreamPart, RetryErrorReason as RetryErrorReason, InferSchema as InferSchema, Job as Job, StreamObjectOnErrorCallback as StreamObjectOnErrorCallback, JSONValue_2 as JSONValue_2, LanguageModelV2CallOptions as LanguageModelV2CallOptions, LanguageModelV2 as LanguageModelV2, EmbeddingModelV2 as EmbeddingModelV2, ImageModelV2 as ImageModelV2, TranscriptionModelV2 as TranscriptionModelV2, SpeechModelV2 as SpeechModelV2, ExtractModelId as ExtractModelId, ExtractLiteralUnion as ExtractLiteralUnion, ProviderV2 as ProviderV2, getOriginalFetch as getOriginalFetch, UIMessageStreamResponseInit as UIMessageStreamResponseInit, InferUIMessageToolCall as InferUIMessageToolCall, Validator as Validator, StandardSchemaV1 as StandardSchemaV1, UIDataTypesToSchemas as UIDataTypesToSchemas, InferUIMessageData as InferUIMessageData, InferUIMessageMetadata as InferUIMessageMetadata, InferUIMessageTools as InferUIMessageTools, Resolvable as Resolvable, FetchFunction as FetchFunction };
|
|
9043
8888
|
export { schemaSymbol, symbol$1, symbol$1_2, symbol$2, symbol$2_2, symbol$3, symbol$3_2, symbol$4, symbol$4_2, symbol$5, symbol$5_2, symbol$6, symbol$6_2, symbol$7, symbol$7_2, symbol$8, symbol$8_2, symbol$9, symbol$9_2, symbol$a, symbol$a_2, symbol$b, symbol$b_2, symbol$c, symbol$c_2, symbol$d, symbol$d_2, symbol$e, symbol, symbol_2, symbol_3, validatorSymbol };
|
package/dist/docs/SKILL.md
CHANGED
|
@@ -3,7 +3,7 @@ name: mastra-voice-murf
|
|
|
3
3
|
description: Documentation for @mastra/voice-murf. Use when working with @mastra/voice-murf APIs, configuration, or implementation.
|
|
4
4
|
metadata:
|
|
5
5
|
package: "@mastra/voice-murf"
|
|
6
|
-
version: "0.12.2
|
|
6
|
+
version: "0.12.2"
|
|
7
7
|
---
|
|
8
8
|
|
|
9
9
|
## When to use
|
|
@@ -20,7 +20,7 @@ export const agent = new Agent({
|
|
|
20
20
|
id: 'voice-agent',
|
|
21
21
|
name: 'Voice Agent',
|
|
22
22
|
instructions: `You are a helpful assistant with both STT and TTS capabilities.`,
|
|
23
|
-
model: 'openai/gpt-5.
|
|
23
|
+
model: 'openai/gpt-5.5',
|
|
24
24
|
voice,
|
|
25
25
|
})
|
|
26
26
|
|
|
@@ -109,7 +109,7 @@ export const agent = new Agent({
|
|
|
109
109
|
id: 'speech-to-speech-agent',
|
|
110
110
|
name: 'Speech-to-Speech Agent',
|
|
111
111
|
instructions: `You are a helpful assistant with speech-to-speech capabilities.`,
|
|
112
|
-
model: 'openai/gpt-5.
|
|
112
|
+
model: 'openai/gpt-5.5',
|
|
113
113
|
tools: {
|
|
114
114
|
// Tools configured on Agent are passed to voice provider
|
|
115
115
|
search,
|
|
@@ -132,6 +132,37 @@ agent.voice.send(microphoneStream)
|
|
|
132
132
|
agent.voice.close()
|
|
133
133
|
```
|
|
134
134
|
|
|
135
|
+
### Per-session voice for concurrent sessions
|
|
136
|
+
|
|
137
|
+
A static `voice` instance is shared across every request. For one-shot text-to-speech this is fine, but realtime and speech-to-speech providers store one WebSocket, one set of tools, and one request context per instance. If you deploy a single agent that handles several live sessions at once, a shared instance lets one session overwrite another session's tools, instructions, and request context.
|
|
138
|
+
|
|
139
|
+
To give each session its own voice, provide `voice` as a resolver. Mastra runs the resolver on every `getVoice()` call and returns a fresh, session-owned instance:
|
|
140
|
+
|
|
141
|
+
```typescript
|
|
142
|
+
import { Agent } from '@mastra/core/agent'
|
|
143
|
+
import { OpenAIRealtimeVoice } from '@mastra/voice-openai-realtime'
|
|
144
|
+
|
|
145
|
+
export const agent = new Agent({
|
|
146
|
+
id: 'support-line',
|
|
147
|
+
name: 'Support Line',
|
|
148
|
+
instructions: ({ requestContext }) => `Help user ${requestContext.get('user')}.`,
|
|
149
|
+
model: 'openai/gpt-5.5',
|
|
150
|
+
voice: ({ requestContext }) => new OpenAIRealtimeVoice({ apiKey: requestContext.get('apiKey') }),
|
|
151
|
+
})
|
|
152
|
+
|
|
153
|
+
// Each concurrent session resolves its own voice instance
|
|
154
|
+
const voice = await agent.getVoice({ requestContext })
|
|
155
|
+
await voice.connect()
|
|
156
|
+
```
|
|
157
|
+
|
|
158
|
+
When you use a resolver:
|
|
159
|
+
|
|
160
|
+
- Each call to `getVoice()` returns a new instance, so concurrent sessions never share state.
|
|
161
|
+
- Mastra does not add tools or instructions to a resolver instance. Configure those inside the resolver or on the provider.
|
|
162
|
+
- You own the lifecycle of the returned instance, so call `disconnect()` or `close()` when the session ends.
|
|
163
|
+
|
|
164
|
+
The `agent.voice` getter has no request context, so it throws when `voice` is a resolver. Use `agent.getVoice({ requestContext })` instead.
|
|
165
|
+
|
|
135
166
|
### Event System
|
|
136
167
|
|
|
137
168
|
The realtime voice provider emits several events you can listen for:
|
|
@@ -209,7 +240,7 @@ export const convertToText = async (input: string | NodeJS.ReadableStream): Prom
|
|
|
209
240
|
export const hybridVoiceAgent = new Agent({
|
|
210
241
|
id: 'hybrid-voice-agent',
|
|
211
242
|
name: 'Hybrid Voice Agent',
|
|
212
|
-
model: 'openai/gpt-5.
|
|
243
|
+
model: 'openai/gpt-5.5',
|
|
213
244
|
instructions: 'You can speak and listen using different providers.',
|
|
214
245
|
voice: new CompositeVoice({
|
|
215
246
|
input: new OpenAIVoice(),
|
|
@@ -221,7 +252,7 @@ export const unifiedVoiceAgent = new Agent({
|
|
|
221
252
|
id: 'unified-voice-agent',
|
|
222
253
|
name: 'Unified Voice Agent',
|
|
223
254
|
instructions: 'You are an agent with both STT and TTS capabilities.',
|
|
224
|
-
model: 'openai/gpt-5.
|
|
255
|
+
model: 'openai/gpt-5.5',
|
|
225
256
|
voice: new OpenAIVoice(),
|
|
226
257
|
})
|
|
227
258
|
|
|
@@ -263,7 +294,7 @@ export const agent = new Agent({
|
|
|
263
294
|
id: 'voice-agent',
|
|
264
295
|
name: 'Voice Agent',
|
|
265
296
|
instructions: `You are a helpful assistant with both STT and TTS capabilities.`,
|
|
266
|
-
model: 'openai/gpt-5.
|
|
297
|
+
model: 'openai/gpt-5.5',
|
|
267
298
|
|
|
268
299
|
// Create a composite voice using OpenAI for listening and PlayAI for speaking
|
|
269
300
|
voice: new CompositeVoice({
|
|
@@ -288,7 +319,7 @@ export const agent = new Agent({
|
|
|
288
319
|
id: 'aisdk-voice-agent',
|
|
289
320
|
name: 'AI SDK Voice Agent',
|
|
290
321
|
instructions: `You are a helpful assistant with voice capabilities.`,
|
|
291
|
-
model: 'openai/gpt-5.
|
|
322
|
+
model: 'openai/gpt-5.5',
|
|
292
323
|
|
|
293
324
|
// Pass AI SDK models directly to CompositeVoice
|
|
294
325
|
voice: new CompositeVoice({
|
|
@@ -16,7 +16,7 @@ const voiceAgent = new Agent({
|
|
|
16
16
|
id: 'voice-agent',
|
|
17
17
|
name: 'Voice Agent',
|
|
18
18
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
19
|
-
model: 'openai/gpt-5.
|
|
19
|
+
model: 'openai/gpt-5.5',
|
|
20
20
|
voice: new OpenAIVoice(),
|
|
21
21
|
})
|
|
22
22
|
```
|
|
@@ -40,7 +40,7 @@ const voiceAgent = new Agent({
|
|
|
40
40
|
id: 'voice-agent',
|
|
41
41
|
name: 'Voice Agent',
|
|
42
42
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
43
|
-
model: 'openai/gpt-5.
|
|
43
|
+
model: 'openai/gpt-5.5',
|
|
44
44
|
voice: new OpenAIVoice(),
|
|
45
45
|
})
|
|
46
46
|
|
|
@@ -68,7 +68,7 @@ const voiceAgent = new Agent({
|
|
|
68
68
|
id: 'voice-agent',
|
|
69
69
|
name: 'Voice Agent',
|
|
70
70
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
71
|
-
model: 'openai/gpt-5.
|
|
71
|
+
model: 'openai/gpt-5.5',
|
|
72
72
|
voice: new AzureVoice(),
|
|
73
73
|
})
|
|
74
74
|
|
|
@@ -95,7 +95,7 @@ const voiceAgent = new Agent({
|
|
|
95
95
|
id: 'voice-agent',
|
|
96
96
|
name: 'Voice Agent',
|
|
97
97
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
98
|
-
model: 'openai/gpt-5.
|
|
98
|
+
model: 'openai/gpt-5.5',
|
|
99
99
|
voice: new ElevenLabsVoice(),
|
|
100
100
|
})
|
|
101
101
|
|
|
@@ -122,7 +122,7 @@ const voiceAgent = new Agent({
|
|
|
122
122
|
id: 'voice-agent',
|
|
123
123
|
name: 'Voice Agent',
|
|
124
124
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
125
|
-
model: 'openai/gpt-5.
|
|
125
|
+
model: 'openai/gpt-5.5',
|
|
126
126
|
voice: new PlayAIVoice(),
|
|
127
127
|
})
|
|
128
128
|
|
|
@@ -149,7 +149,7 @@ const voiceAgent = new Agent({
|
|
|
149
149
|
id: 'voice-agent',
|
|
150
150
|
name: 'Voice Agent',
|
|
151
151
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
152
|
-
model: 'openai/gpt-5.
|
|
152
|
+
model: 'openai/gpt-5.5',
|
|
153
153
|
voice: new GoogleVoice(),
|
|
154
154
|
})
|
|
155
155
|
|
|
@@ -176,7 +176,7 @@ const voiceAgent = new Agent({
|
|
|
176
176
|
id: 'voice-agent',
|
|
177
177
|
name: 'Voice Agent',
|
|
178
178
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
179
|
-
model: 'openai/gpt-5.
|
|
179
|
+
model: 'openai/gpt-5.5',
|
|
180
180
|
voice: new CloudflareVoice(),
|
|
181
181
|
})
|
|
182
182
|
|
|
@@ -203,7 +203,7 @@ const voiceAgent = new Agent({
|
|
|
203
203
|
id: 'voice-agent',
|
|
204
204
|
name: 'Voice Agent',
|
|
205
205
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
206
|
-
model: 'openai/gpt-5.
|
|
206
|
+
model: 'openai/gpt-5.5',
|
|
207
207
|
voice: new DeepgramVoice(),
|
|
208
208
|
})
|
|
209
209
|
|
|
@@ -230,7 +230,7 @@ const voiceAgent = new Agent({
|
|
|
230
230
|
id: 'voice-agent',
|
|
231
231
|
name: 'Voice Agent',
|
|
232
232
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
233
|
-
model: 'openai/gpt-5.
|
|
233
|
+
model: 'openai/gpt-5.5',
|
|
234
234
|
voice: new InworldVoice(),
|
|
235
235
|
})
|
|
236
236
|
|
|
@@ -257,7 +257,7 @@ const voiceAgent = new Agent({
|
|
|
257
257
|
id: 'voice-agent',
|
|
258
258
|
name: 'Voice Agent',
|
|
259
259
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
260
|
-
model: 'openai/gpt-5.
|
|
260
|
+
model: 'openai/gpt-5.5',
|
|
261
261
|
voice: new SpeechifyVoice(),
|
|
262
262
|
})
|
|
263
263
|
|
|
@@ -284,7 +284,7 @@ const voiceAgent = new Agent({
|
|
|
284
284
|
id: 'voice-agent',
|
|
285
285
|
name: 'Voice Agent',
|
|
286
286
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
287
|
-
model: 'openai/gpt-5.
|
|
287
|
+
model: 'openai/gpt-5.5',
|
|
288
288
|
voice: new SarvamVoice(),
|
|
289
289
|
})
|
|
290
290
|
|
|
@@ -311,7 +311,7 @@ const voiceAgent = new Agent({
|
|
|
311
311
|
id: 'voice-agent',
|
|
312
312
|
name: 'Voice Agent',
|
|
313
313
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
314
|
-
model: 'openai/gpt-5.
|
|
314
|
+
model: 'openai/gpt-5.5',
|
|
315
315
|
voice: new MurfVoice(),
|
|
316
316
|
})
|
|
317
317
|
|
|
@@ -346,7 +346,7 @@ const voiceAgent = new Agent({
|
|
|
346
346
|
id: 'voice-agent',
|
|
347
347
|
name: 'Voice Agent',
|
|
348
348
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
349
|
-
model: 'openai/gpt-5.
|
|
349
|
+
model: 'openai/gpt-5.5',
|
|
350
350
|
voice: new OpenAIVoice(),
|
|
351
351
|
})
|
|
352
352
|
|
|
@@ -375,7 +375,7 @@ const voiceAgent = new Agent({
|
|
|
375
375
|
id: 'voice-agent',
|
|
376
376
|
name: 'Voice Agent',
|
|
377
377
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
378
|
-
model: 'openai/gpt-5.
|
|
378
|
+
model: 'openai/gpt-5.5',
|
|
379
379
|
voice: new AzureVoice(),
|
|
380
380
|
})
|
|
381
381
|
|
|
@@ -403,7 +403,7 @@ const voiceAgent = new Agent({
|
|
|
403
403
|
id: 'voice-agent',
|
|
404
404
|
name: 'Voice Agent',
|
|
405
405
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
406
|
-
model: 'openai/gpt-5.
|
|
406
|
+
model: 'openai/gpt-5.5',
|
|
407
407
|
voice: new ElevenLabsVoice(),
|
|
408
408
|
})
|
|
409
409
|
|
|
@@ -431,7 +431,7 @@ const voiceAgent = new Agent({
|
|
|
431
431
|
id: 'voice-agent',
|
|
432
432
|
name: 'Voice Agent',
|
|
433
433
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
434
|
-
model: 'openai/gpt-5.
|
|
434
|
+
model: 'openai/gpt-5.5',
|
|
435
435
|
voice: new GoogleVoice(),
|
|
436
436
|
})
|
|
437
437
|
|
|
@@ -459,7 +459,7 @@ const voiceAgent = new Agent({
|
|
|
459
459
|
id: 'voice-agent',
|
|
460
460
|
name: 'Voice Agent',
|
|
461
461
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
462
|
-
model: 'openai/gpt-5.
|
|
462
|
+
model: 'openai/gpt-5.5',
|
|
463
463
|
voice: new CloudflareVoice(),
|
|
464
464
|
})
|
|
465
465
|
|
|
@@ -487,7 +487,7 @@ const voiceAgent = new Agent({
|
|
|
487
487
|
id: 'voice-agent',
|
|
488
488
|
name: 'Voice Agent',
|
|
489
489
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
490
|
-
model: 'openai/gpt-5.
|
|
490
|
+
model: 'openai/gpt-5.5',
|
|
491
491
|
voice: new DeepgramVoice(),
|
|
492
492
|
})
|
|
493
493
|
|
|
@@ -515,7 +515,7 @@ const voiceAgent = new Agent({
|
|
|
515
515
|
id: 'voice-agent',
|
|
516
516
|
name: 'Voice Agent',
|
|
517
517
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
518
|
-
model: 'openai/gpt-5.
|
|
518
|
+
model: 'openai/gpt-5.5',
|
|
519
519
|
voice: new InworldVoice(),
|
|
520
520
|
})
|
|
521
521
|
|
|
@@ -543,7 +543,7 @@ const voiceAgent = new Agent({
|
|
|
543
543
|
id: 'voice-agent',
|
|
544
544
|
name: 'Voice Agent',
|
|
545
545
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
546
|
-
model: 'openai/gpt-5.
|
|
546
|
+
model: 'openai/gpt-5.5',
|
|
547
547
|
voice: new SarvamVoice(),
|
|
548
548
|
})
|
|
549
549
|
|
|
@@ -575,7 +575,7 @@ const voiceAgent = new Agent({
|
|
|
575
575
|
id: 'voice-agent',
|
|
576
576
|
name: 'Voice Agent',
|
|
577
577
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
578
|
-
model: 'openai/gpt-5.
|
|
578
|
+
model: 'openai/gpt-5.5',
|
|
579
579
|
voice: new OpenAIRealtimeVoice(),
|
|
580
580
|
})
|
|
581
581
|
|
|
@@ -605,7 +605,7 @@ const voiceAgent = new Agent({
|
|
|
605
605
|
id: 'voice-agent',
|
|
606
606
|
name: 'Voice Agent',
|
|
607
607
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
608
|
-
model: 'openai/gpt-5.
|
|
608
|
+
model: 'openai/gpt-5.5',
|
|
609
609
|
voice: new GeminiLiveVoice({
|
|
610
610
|
// Live API mode
|
|
611
611
|
apiKey: process.env.GOOGLE_API_KEY,
|
|
@@ -654,7 +654,7 @@ const voiceAgent = new Agent({
|
|
|
654
654
|
id: 'voice-agent',
|
|
655
655
|
name: 'Voice Agent',
|
|
656
656
|
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
657
|
-
model: 'openai/gpt-5.
|
|
657
|
+
model: 'openai/gpt-5.5',
|
|
658
658
|
voice: new NovaSonicVoice({
|
|
659
659
|
region: 'us-east-1',
|
|
660
660
|
speaker: 'matthew',
|
|
@@ -686,6 +686,48 @@ await voiceAgent.voice.send(micStream)
|
|
|
686
686
|
|
|
687
687
|
Visit the [AWS Nova Sonic Reference](https://mastra.ai/reference/voice/aws-nova-sonic) for more information on the AWS Nova Sonic voice provider.
|
|
688
688
|
|
|
689
|
+
**Inworld Realtime**:
|
|
690
|
+
|
|
691
|
+
```typescript
|
|
692
|
+
import { Agent } from '@mastra/core/agent'
|
|
693
|
+
import { playAudio, getMicrophoneStream } from '@mastra/node-audio'
|
|
694
|
+
import { InworldRealtimeVoice } from '@mastra/voice-inworld'
|
|
695
|
+
|
|
696
|
+
const voiceAgent = new Agent({
|
|
697
|
+
id: 'voice-agent',
|
|
698
|
+
name: 'Voice Agent',
|
|
699
|
+
instructions: 'You are a voice assistant that can help users with their tasks.',
|
|
700
|
+
model: 'openai/gpt-5.5',
|
|
701
|
+
voice: new InworldRealtimeVoice({
|
|
702
|
+
apiKey: process.env.INWORLD_API_KEY,
|
|
703
|
+
model: 'inworld/models/gemma-4-26b-a4b-it',
|
|
704
|
+
speaker: 'Sarah',
|
|
705
|
+
}),
|
|
706
|
+
})
|
|
707
|
+
|
|
708
|
+
// Connect before using speak/send
|
|
709
|
+
await voiceAgent.voice.connect()
|
|
710
|
+
|
|
711
|
+
// Listen for agent audio (PCM stream)
|
|
712
|
+
voiceAgent.voice.on('speaker', stream => {
|
|
713
|
+
playAudio(stream)
|
|
714
|
+
})
|
|
715
|
+
|
|
716
|
+
// Listen for text responses and transcriptions
|
|
717
|
+
voiceAgent.voice.on('writing', ({ text, role }) => {
|
|
718
|
+
console.log(`${role}: ${text}`)
|
|
719
|
+
})
|
|
720
|
+
|
|
721
|
+
// Initiate the conversation
|
|
722
|
+
await voiceAgent.voice.speak('How can I help you today?')
|
|
723
|
+
|
|
724
|
+
// Send continuous audio from the microphone
|
|
725
|
+
const micStream = getMicrophoneStream()
|
|
726
|
+
await voiceAgent.voice.send(micStream)
|
|
727
|
+
```
|
|
728
|
+
|
|
729
|
+
Visit the [Inworld Realtime Reference](https://mastra.ai/reference/voice/inworld-realtime) for more information on the Inworld Realtime voice provider.
|
|
730
|
+
|
|
689
731
|
**xAI**:
|
|
690
732
|
|
|
691
733
|
```typescript
|
|
@@ -1051,6 +1093,26 @@ const voice = new NovaSonicVoice({
|
|
|
1051
1093
|
|
|
1052
1094
|
Visit the [AWS Nova Sonic Reference](https://mastra.ai/reference/voice/aws-nova-sonic) for more information on the AWS Nova Sonic voice provider.
|
|
1053
1095
|
|
|
1096
|
+
**Inworld Realtime**:
|
|
1097
|
+
|
|
1098
|
+
```typescript
|
|
1099
|
+
// Inworld Realtime Voice Configuration
|
|
1100
|
+
const voice = new InworldRealtimeVoice({
|
|
1101
|
+
apiKey: process.env.INWORLD_API_KEY,
|
|
1102
|
+
model: 'inworld/models/gemma-4-26b-a4b-it',
|
|
1103
|
+
speaker: 'Sarah',
|
|
1104
|
+
// Typed Inworld realtime knobs (semantic VAD, playback speed, MCP tool routing, ...)
|
|
1105
|
+
session: {
|
|
1106
|
+
audio: {
|
|
1107
|
+
output: { speed: 1.1 },
|
|
1108
|
+
input: { turn_detection: { type: 'semantic_vad', eagerness: 'high' } },
|
|
1109
|
+
},
|
|
1110
|
+
},
|
|
1111
|
+
})
|
|
1112
|
+
```
|
|
1113
|
+
|
|
1114
|
+
Visit the [Inworld Realtime Reference](https://mastra.ai/reference/voice/inworld-realtime) for more information on the Inworld Realtime voice provider.
|
|
1115
|
+
|
|
1054
1116
|
**AI SDK**:
|
|
1055
1117
|
|
|
1056
1118
|
```typescript
|
|
@@ -1070,7 +1132,7 @@ const voiceAgent = new Agent({
|
|
|
1070
1132
|
id: 'aisdk-voice-agent',
|
|
1071
1133
|
name: 'AI SDK Voice Agent',
|
|
1072
1134
|
instructions: 'You are a helpful assistant with voice capabilities.',
|
|
1073
|
-
model: 'openai/gpt-5.
|
|
1135
|
+
model: 'openai/gpt-5.5',
|
|
1074
1136
|
voice,
|
|
1075
1137
|
})
|
|
1076
1138
|
```
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mastra/voice-murf",
|
|
3
|
-
"version": "0.12.2
|
|
3
|
+
"version": "0.12.2",
|
|
4
4
|
"description": "Mastra Murf voice integration",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"files": [
|
|
@@ -31,9 +31,9 @@
|
|
|
31
31
|
"tsup": "^8.5.1",
|
|
32
32
|
"typescript": "^6.0.3",
|
|
33
33
|
"vitest": "4.1.5",
|
|
34
|
-
"@internal/
|
|
35
|
-
"@internal/
|
|
36
|
-
"@internal/
|
|
34
|
+
"@internal/lint": "0.0.100",
|
|
35
|
+
"@internal/voice": "0.0.1",
|
|
36
|
+
"@internal/types-builder": "0.0.75"
|
|
37
37
|
},
|
|
38
38
|
"peerDependencies": {
|
|
39
39
|
"zod": "^3.25.0 || ^4.0.0"
|