@deepseek-ai/dsh-experimental-api-speech-to-text 0.1.7-alpha.2 → 0.1.7-rc.2

package/README.i18n.yaml

@@ -1,6 +1,37 @@
-# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
-# side as of the last confirmed-consistent state. Both languages carry equal authority;
-# after editing either side, bring the other along and re-record with:
+# Bilingual-pair consistency record for README.md (docs/i18n/README.md): per heading
+# section, a hash of its English and Chinese blocks outside code blocks and generated regions.
+# After editing either side, bring the other along and re-record with:
 #   pnpm run verify-translation-pairing --write packages/experimental/api-speech-to-text/README.md
-README.md: ecc83c15b4217323c074733ae77259261598803c
-README.zh.md: 46ac6a16c902c32f2effb23a98141b1afbdccd37
+/:
+  en: 492c18403fe67bc0
+  zh: 547cb261324c7691
+/deepseek-ai-dsh-experimental-api-speech-to-text:
+  en: 9273fdafb9804346
+  zh: 4fec3a81b4d29fc6
+/deepseek-ai-dsh-experimental-api-speech-to-text/summary:
+  en: 81978b4b6c035754
+  zh: f1a08200eb31c020
+/deepseek-ai-dsh-experimental-api-speech-to-text/table-of-contents:
+  en: d152484eb41ac6b4
+  zh: 364c80ee00a374ff
+/deepseek-ai-dsh-experimental-api-speech-to-text/use-this-package:
+  en: 23efd2789fe70cd9
+  zh: 3edd559ee5975742
+/deepseek-ai-dsh-experimental-api-speech-to-text/understand-the-implementation:
+  en: 83c580c412f4026a
+  zh: cd6742fef27de9e5
+/deepseek-ai-dsh-experimental-api-speech-to-text/further-exploration:
+  en: 5c64426add9d79d6
+  zh: 4e06aa2ead8d2ccd
+/deepseek-ai-dsh-experimental-api-speech-to-text/model-experience:
+  en: 3c5a85f8b7129ce8
+  zh: d5f491ca88d9a810
+/deepseek-ai-dsh-experimental-api-speech-to-text/model-experience/kv-cache-effect:
+  en: b5ec79db2e1d0dbc
+  zh: e5aff567ae5ee6fc
+/deepseek-ai-dsh-experimental-api-speech-to-text/known-limitations-and-deferred-work:
+  en: 067a5c6a1772f2eb
+  zh: 6973c5a4a950c5cf
+/deepseek-ai-dsh-experimental-api-speech-to-text/known-limitations-and-deferred-work/dev-note:
+  en: 0225ba087c5266f3
+  zh: 4e706f1b2b984e63

package/README.md

@@ -27,6 +27,8 @@ The `speech` Remote connects browser recordings to `ctx.speechToText`. It expose
 
 Compose with the speech Service Definition and Typert. `maxAudioBytes` and `maxDurationSeconds` limit accepted recordings. The browser UI mounts this package’s generated `/remote` contribution when enabled.
 
+`prepare(providerId, { downloadSource })` forwards one advertised source to the provider; omission keeps its configured policy. The catalog carries `downloadSources` for the picker. The provider rejects unavailable choices and changes to an active task’s source.
+
 -----
 
 <a id="understand-the-implementation"></a>

package/README.zh.md

@@ -27,6 +27,8 @@ kind: "package-reference"
 
 与语音服务定义和 Typert 组合。`maxAudioBytes` 与 `maxDurationSeconds` 限制接收的录音。浏览器 UI 启用时挂载本包生成的 `/remote` 贡献。
 
+`prepare(providerId, { downloadSource })` 将公布的一个下载源转发给提供方，省略时保留其配置策略。目录通过 `downloadSources` 提供选择器选项。提供方拒绝不可用的源，以及对进行中任务的源变更。
+
 -----
 
 <a id="understand-the-implementation"></a>

package/lib/index.js

@@ -173,9 +173,10 @@ let SpeechController = (() => {
 		/**
 		* Start or join one Host-owned preparation task.
 		* @param providerId - selected recognizer.
+		* @param options - task-local source selection validated by the provider.
 		*/
-		prepare(providerId) {
-			this.ctx.speechToText.prepare(providerId);
+		prepare(providerId, options) {
+			this.ctx.speechToText.prepare(providerId, options);
 		}
 		/**
 		* Explicitly cancel resource preparation.

package/lib/typert.host.js

@@ -69,6 +69,7 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_catalog_result$sch
   'minimumMinutes': z.number().readonly(),
   'maximumMinutes': z.number().readonly(),
 }).readonly().optional(),
+  'downloadSources': z.array(z.string()).readonly().optional(),
 })).readonly(),
   'selection': z.object({
   'providerId': z.intersection(z.string(), z.unknown()).readonly(),
@@ -146,6 +147,7 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_follow_result$sche
   'minimumMinutes': z.number().readonly(),
   'maximumMinutes': z.number().readonly(),
 }).readonly().optional(),
+  'downloadSources': z.array(z.string()).readonly().optional(),
 })).readonly(),
   'selection': z.object({
   'providerId': z.intersection(z.string(), z.unknown()).readonly(),
@@ -154,6 +156,10 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_follow_result$sche
 }))
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema$value
 const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema$value ??= z.intersection(z.string(), z.unknown()))
+let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema$value
+const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema$value ??= z.union([z.undefined(), z.object({
+  'downloadSource': z.string().readonly().optional(),
+})]))
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema$value
 const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema$value ??= z.void())
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_transcribe_parameter_0$schema$value
@@ -198,7 +204,7 @@ export const TYPERT = {
         typeSymbol: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/cancelPreparation:result',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_cancelPreparation_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":79,"column":3},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":80,"column":3},
     },
     {
       id: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/catalog',
@@ -274,13 +280,24 @@ export const TYPERT = {
             create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema,
           },
         },
+        {
+          name: 'options',
+          wire: 'options',
+          source: 'json',
+          acceptsUndefined: true,
+          codec: {
+            mode: 'strict',
+            typeSymbol: '@deepseek-ai/dsh-experimental-speech-to-text/types#SpeechPreparationOptions',
+            create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema,
+          },
+        },
       ],
       result: {
         mode: 'strict',
         typeSymbol: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/prepare:result',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":71,"column":3},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":72,"column":3},
     },
     {
       id: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/transcribe',
@@ -306,7 +323,7 @@ export const TYPERT = {
         typeSymbol: '@deepseek-ai/dsh-experimental-speech-to-text/types#Transcript',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_transcribe_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":88,"column":9},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":89,"column":9},
     },
   ],
   model: {
@@ -343,9 +360,9 @@ export const TYPERT = {
           {
             "kind": "method",
             "name": "prepare",
-            "signature": "@Remote prepare(providerId: SpeechProviderId): void",
+            "signature": "@Remote prepare(providerId: SpeechProviderId, options?: SpeechPreparationOptions): void",
             "summary": "Start or join one Host-owned preparation task.",
-            "jsDoc": "/**\n * Start or join one Host-owned preparation task.\n * @param providerId - selected recognizer.\n */"
+            "jsDoc": "/**\n * Start or join one Host-owned preparation task.\n * @param providerId - selected recognizer.\n * @param options - task-local source selection validated by the provider.\n */"
           },
           {
             "kind": "method",
@@ -375,6 +392,10 @@ export const TYPERT = {
             "name": "SpeechDownloadFailure",
             "declaration": "export interface SpeechDownloadFailure {\n    readonly resource: string;\n    readonly source: string;\n    readonly reason: 'network' | 'dns' | 'timeout' | 'certificate' | 'http' | 'integrity' | 'storage' | 'unknown';\n    readonly code?: string;\n    readonly status?: number;\n}"
           },
+          {
+            "name": "SpeechPreparationOptions",
+            "declaration": "export interface SpeechPreparationOptions {\n    readonly downloadSource?: string;\n}"
+          },
           {
             "name": "SpeechPreparationState",
             "declaration": "export type SpeechPreparationState = ({ readonly phase: 'unprepared' | 'ready' | 'standby' | 'cancelled'; } | { readonly phase: 'downloading'; readonly resource: string; readonly completedBytes: number; readonly totalBytes?: number; } | { readonly phase: 'checking' | 'loading' | 'waking' | 'cancelling'; readonly startedAt: number; } | { readonly phase: 'failed'; readonly message: string; readonly download?: SpeechDownloadFailure; }) & { readonly step?: SpeechPreparationStepKind; readonly steps?: readonly SpeechPreparationStep[]; };"
@@ -393,7 +414,7 @@ export const TYPERT = {
           },
           {
             "name": "SpeechProviderInfo",
-            "declaration": "export interface SpeechProviderInfo {\n    readonly id: SpeechProviderId;\n    readonly name: string;\n    readonly location: 'host-local' | 'cloud';\n    readonly languages: readonly string[];\n    readonly setupEstimate?: SpeechSetupEstimate;\n}"
+            "declaration": "export interface SpeechProviderInfo {\n    readonly id: SpeechProviderId;\n    readonly name: string;\n    readonly location: 'host-local' | 'cloud';\n    readonly languages: readonly string[];\n    readonly setupEstimate?: SpeechSetupEstimate;\n    readonly downloadSources?: readonly string[];\n}"
           },
           {
             "name": "SpeechProviderView",

package/lib/typert.remote-client.d.ts

@@ -5,7 +5,7 @@ import type {
   TypertRemoteContribution,
 } from '@deepseek-ai/dsh-typert-protocol'
 import type { SpeechCatalog, TranscriptionRequest } from '@deepseek-ai/dsh-experimental-api-speech-to-text/types'
-import type { SpeechProviderId, SpeechSelectionPatch, Transcript } from '@deepseek-ai/dsh-experimental-speech-to-text/types'
+import type { SpeechPreparationOptions, SpeechProviderId, SpeechSelectionPatch, Transcript } from '@deepseek-ai/dsh-experimental-speech-to-text/types'
 
 declare module '@deepseek-ai/dsh-typert-protocol' {
   interface TypertRemoteNamespace$737065656368 {
@@ -13,7 +13,7 @@ declare module '@deepseek-ai/dsh-typert-protocol' {
     catalog: () => Promise<RemoteResult<SpeechCatalog>>
     configure: (patch: SpeechSelectionPatch) => Promise<RemoteResult<void>>
     follow: (signal?: AbortSignal) => RemoteStreamHandle<SpeechCatalog, never>
-    prepare: (providerId: SpeechProviderId) => Promise<RemoteResult<void>>
+    prepare: (providerId: SpeechProviderId, options?: SpeechPreparationOptions) => Promise<RemoteResult<void>>
     transcribe: (request: TranscriptionRequest, signal?: AbortSignal) => Promise<RemoteResult<Transcript>>
   }
   interface TypertRemoteMap {
@@ -21,7 +21,7 @@ declare module '@deepseek-ai/dsh-typert-protocol' {
     'speech/catalog': () => Promise<RemoteResult<SpeechCatalog>>
     'speech/configure': (patch: SpeechSelectionPatch) => Promise<RemoteResult<void>>
     'speech/follow': (signal?: AbortSignal) => RemoteStreamHandle<SpeechCatalog, never>
-    'speech/prepare': (providerId: SpeechProviderId) => Promise<RemoteResult<void>>
+    'speech/prepare': (providerId: SpeechProviderId, options?: SpeechPreparationOptions) => Promise<RemoteResult<void>>
     'speech/transcribe': (request: TranscriptionRequest, signal?: AbortSignal) => Promise<RemoteResult<Transcript>>
   }
   interface TypertRemoteNamespaceMap {

package/lib/typert.remote-client.js

@@ -69,6 +69,7 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_catalog_result$sch
   'minimumMinutes': z.number().readonly(),
   'maximumMinutes': z.number().readonly(),
 }).readonly().optional(),
+  'downloadSources': z.array(z.string()).readonly().optional(),
 })).readonly(),
   'selection': z.object({
   'providerId': z.intersection(z.string(), z.unknown()).readonly(),
@@ -146,6 +147,7 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_follow_result$sche
   'minimumMinutes': z.number().readonly(),
   'maximumMinutes': z.number().readonly(),
 }).readonly().optional(),
+  'downloadSources': z.array(z.string()).readonly().optional(),
 })).readonly(),
   'selection': z.object({
   'providerId': z.intersection(z.string(), z.unknown()).readonly(),
@@ -154,6 +156,10 @@ const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_follow_result$sche
 }))
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema$value
 const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema$value ??= z.intersection(z.string(), z.unknown()))
+let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema$value
+const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema$value ??= z.union([z.undefined(), z.object({
+  'downloadSource': z.string().readonly().optional(),
+})]))
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema$value
 const _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema = () => (_deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema$value ??= z.void())
 let _deepseek_ai_dsh_experimental_api_speech_to_text_speech_transcribe_parameter_0$schema$value
@@ -195,7 +201,7 @@ export const TYPERT_REMOTE = {
         typeSymbol: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/cancelPreparation:result',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_cancelPreparation_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":79,"column":3},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":80,"column":3},
     },
     {
       id: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/catalog',
@@ -271,13 +277,24 @@ export const TYPERT_REMOTE = {
             create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_0$schema,
           },
         },
+        {
+          name: 'options',
+          wire: 'options',
+          source: 'json',
+          acceptsUndefined: true,
+          codec: {
+            mode: 'strict',
+            typeSymbol: '@deepseek-ai/dsh-experimental-speech-to-text/types#SpeechPreparationOptions',
+            create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_parameter_1$schema,
+          },
+        },
       ],
       result: {
         mode: 'strict',
         typeSymbol: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/prepare:result',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_prepare_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":71,"column":3},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":72,"column":3},
     },
     {
       id: '@deepseek-ai/dsh-experimental-api-speech-to-text#speech/transcribe',
@@ -303,7 +320,7 @@ export const TYPERT_REMOTE = {
         typeSymbol: '@deepseek-ai/dsh-experimental-speech-to-text/types#Transcript',
         create: _deepseek_ai_dsh_experimental_api_speech_to_text_speech_transcribe_result$schema,
       },
-      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":88,"column":9},
+      sourceLocation: {"file":"packages/experimental/api-speech-to-text/src/index.ts","line":89,"column":9},
     },
   ],
 }

package/lib/types/index.d.ts

@@ -2,7 +2,7 @@
 import { Context } from '@deepseek-ai/cordis';
 import z from '@deepseek-ai/schemastery';
 import { TypertRemoteService } from '@deepseek-ai/dsh-typert-protocol';
-import type { SpeechProviderId, SpeechSelectionPatch, Transcript } from '@deepseek-ai/dsh-experimental-speech-to-text/types';
+import type { SpeechPreparationOptions, SpeechProviderId, SpeechSelectionPatch, Transcript } from '@deepseek-ai/dsh-experimental-speech-to-text/types';
 import type { SpeechCatalog, TranscriptionRequest } from './types.ts';
 export type * from './types.ts';
 declare module '@deepseek-ai/cordis' {
@@ -44,8 +44,9 @@ export default class SpeechController extends TypertRemoteService {
     /**
      * Start or join one Host-owned preparation task.
      * @param providerId - selected recognizer.
+     * @param options - task-local source selection validated by the provider.
      */
-    prepare(providerId: SpeechProviderId): void;
+    prepare(providerId: SpeechProviderId, options?: SpeechPreparationOptions): void;
     /**
      * Explicitly cancel resource preparation.
      * @param providerId - selected recognizer.

package/lib/types/index.js

@@ -96,8 +96,9 @@ let SpeechController = (() => {
         /**
          * Start or join one Host-owned preparation task.
          * @param providerId - selected recognizer.
+         * @param options - task-local source selection validated by the provider.
          */
-        prepare(providerId) { this.ctx.speechToText.prepare(providerId); }
+        prepare(providerId, options) { this.ctx.speechToText.prepare(providerId, options); }
         /**
          * Explicitly cancel resource preparation.
          * @param providerId - selected recognizer.

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@deepseek-ai/dsh-experimental-api-speech-to-text",
   "description": "Authenticated experimental speech transcription for browser clients",
-  "version": "0.1.7-alpha.2",
+  "version": "0.1.7-rc.2",
   "publishConfig": {
     "access": "public"
   },
@@ -49,12 +49,12 @@
   },
   "peerDependencies": {
     "@deepseek-ai/cordis": "~4.0.4",
-    "@deepseek-ai/dsh-experimental-speech-to-text": "0.1.7-alpha.2",
-    "@deepseek-ai/dsh-typert-protocol": "0.1.7-alpha.2"
+    "@deepseek-ai/dsh-experimental-speech-to-text": "0.1.7-rc.2",
+    "@deepseek-ai/dsh-typert-protocol": "0.1.7-rc.2"
   },
   "devDependencies": {
     "@deepseek-ai/cordis": "~4.0.4",
-    "@deepseek-ai/dsh-experimental-speech-to-text": "0.1.7-alpha.2",
-    "@deepseek-ai/dsh-typert-protocol": "0.1.7-alpha.2"
+    "@deepseek-ai/dsh-typert-protocol": "0.1.7-rc.2",
+    "@deepseek-ai/dsh-experimental-speech-to-text": "0.1.7-rc.2"
   }
 }