mirror of
https://github.com/openclaw/openclaw.git
synced 2026-05-06 09:00:42 +00:00
fix(plugin-sdk): split tts runtime contract types
This commit is contained in:
Vincent Koc
@@ -1,151 +1,21 @@
|
||||
import type { ReplyPayload } from "../auto-reply/reply-payload.js";
|
||||
import type { OpenClawConfig } from "../config/types.openclaw.js";
|
||||
import type { TtsAutoMode, TtsProvider } from "../config/types.tts.js";
|
||||
import type {
|
||||
SpeechProviderConfig,
|
||||
SpeechVoiceOption,
|
||||
TtsDirectiveOverrides,
|
||||
TtsDirectiveParseResult,
|
||||
} from "../tts/provider-types.js";
|
||||
import type { ResolvedTtsConfig, ResolvedTtsModelOverrides } from "../tts/tts-types.js";
|
||||
import {
|
||||
createLazyFacadeObjectValue,
|
||||
loadActivatedBundledPluginPublicSurfaceModuleSync,
|
||||
} from "./facade-runtime.js";
|
||||
import type {
|
||||
ResolvedTtsConfig,
|
||||
ResolvedTtsModelOverrides,
|
||||
TtsDirectiveOverrides,
|
||||
TtsDirectiveParseResult,
|
||||
TtsResult,
|
||||
TtsRuntimeFacade,
|
||||
TtsSynthesisResult,
|
||||
TtsTelephonyResult,
|
||||
} from "./tts-runtime.types.js";
|
||||
|
||||
// Manual facade. Keep loader boundary explicit and avoid typing this public SDK
|
||||
// seam through the bundled speech-core runtime surface.
|
||||
type TtsAttemptReasonCode =
|
||||
| "success"
|
||||
| "no_provider_registered"
|
||||
| "not_configured"
|
||||
| "unsupported_for_telephony"
|
||||
| "timeout"
|
||||
| "provider_error";
|
||||
|
||||
type TtsProviderAttempt = {
|
||||
provider: string;
|
||||
outcome: "success" | "skipped" | "failed";
|
||||
reasonCode: TtsAttemptReasonCode;
|
||||
latencyMs?: number;
|
||||
error?: string;
|
||||
};
|
||||
|
||||
type TtsStatusEntry = {
|
||||
timestamp: number;
|
||||
success: boolean;
|
||||
textLength: number;
|
||||
summarized: boolean;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
latencyMs?: number;
|
||||
error?: string;
|
||||
};
|
||||
|
||||
type SummarizeResult = {
|
||||
summary: string;
|
||||
latencyMs: number;
|
||||
inputLength: number;
|
||||
outputLength: number;
|
||||
};
|
||||
|
||||
type ResolveTtsAutoModeParams = {
|
||||
config: ResolvedTtsConfig;
|
||||
prefsPath: string;
|
||||
sessionAuto?: string;
|
||||
};
|
||||
|
||||
type ResolveExplicitTtsOverridesParams = {
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
provider?: string;
|
||||
modelId?: string;
|
||||
voiceId?: string;
|
||||
};
|
||||
|
||||
type TtsRequestParams = {
|
||||
text: string;
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
channel?: string;
|
||||
overrides?: TtsDirectiveOverrides;
|
||||
disableFallback?: boolean;
|
||||
};
|
||||
|
||||
type TtsTelephonyRequestParams = {
|
||||
text: string;
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
};
|
||||
|
||||
type ListSpeechVoicesParams = {
|
||||
provider: string;
|
||||
cfg?: OpenClawConfig;
|
||||
config?: ResolvedTtsConfig;
|
||||
apiKey?: string;
|
||||
baseUrl?: string;
|
||||
};
|
||||
|
||||
type MaybeApplyTtsToPayloadParams = {
|
||||
payload: ReplyPayload;
|
||||
cfg: OpenClawConfig;
|
||||
channel?: string;
|
||||
kind?: "tool" | "block" | "final";
|
||||
inboundAudio?: boolean;
|
||||
ttsAuto?: string;
|
||||
};
|
||||
|
||||
type TtsTestFacade = {
|
||||
parseTtsDirectives: (...args: unknown[]) => TtsDirectiveParseResult;
|
||||
resolveModelOverridePolicy: (...args: unknown[]) => ResolvedTtsModelOverrides;
|
||||
supportsNativeVoiceNoteTts: (channel: string | undefined) => boolean;
|
||||
summarizeText: (...args: unknown[]) => Promise<SummarizeResult>;
|
||||
getResolvedSpeechProviderConfig: (
|
||||
config: ResolvedTtsConfig,
|
||||
providerId: string,
|
||||
cfg?: OpenClawConfig,
|
||||
) => SpeechProviderConfig;
|
||||
formatTtsProviderError: (provider: TtsProvider, err: unknown) => string;
|
||||
sanitizeTtsErrorForLog: (err: unknown) => string;
|
||||
};
|
||||
|
||||
type FacadeModule = {
|
||||
_test: TtsTestFacade;
|
||||
buildTtsSystemPromptHint: (cfg: OpenClawConfig) => string | undefined;
|
||||
getLastTtsAttempt: () => TtsStatusEntry | undefined;
|
||||
getResolvedSpeechProviderConfig: (
|
||||
config: ResolvedTtsConfig,
|
||||
providerId: string,
|
||||
cfg?: OpenClawConfig,
|
||||
) => SpeechProviderConfig;
|
||||
getTtsMaxLength: (prefsPath: string) => number;
|
||||
getTtsProvider: (config: ResolvedTtsConfig, prefsPath: string) => TtsProvider;
|
||||
isSummarizationEnabled: (prefsPath: string) => boolean;
|
||||
isTtsEnabled: (config: ResolvedTtsConfig, prefsPath: string, sessionAuto?: string) => boolean;
|
||||
isTtsProviderConfigured: (
|
||||
config: ResolvedTtsConfig,
|
||||
provider: TtsProvider,
|
||||
cfg?: OpenClawConfig,
|
||||
) => boolean;
|
||||
listSpeechVoices: (params: ListSpeechVoicesParams) => Promise<SpeechVoiceOption[]>;
|
||||
maybeApplyTtsToPayload: (params: MaybeApplyTtsToPayloadParams) => Promise<ReplyPayload>;
|
||||
resolveExplicitTtsOverrides: (params: ResolveExplicitTtsOverridesParams) => TtsDirectiveOverrides;
|
||||
resolveTtsAutoMode: (params: ResolveTtsAutoModeParams) => TtsAutoMode;
|
||||
resolveTtsConfig: (cfg: OpenClawConfig) => ResolvedTtsConfig;
|
||||
resolveTtsPrefsPath: (config: ResolvedTtsConfig) => string;
|
||||
resolveTtsProviderOrder: (primary: TtsProvider, cfg?: OpenClawConfig) => TtsProvider[];
|
||||
setLastTtsAttempt: (entry: TtsStatusEntry | undefined) => void;
|
||||
setSummarizationEnabled: (prefsPath: string, enabled: boolean) => void;
|
||||
setTtsAutoMode: (prefsPath: string, mode: TtsAutoMode) => void;
|
||||
setTtsEnabled: (prefsPath: string, enabled: boolean) => void;
|
||||
setTtsMaxLength: (prefsPath: string, maxLength: number) => void;
|
||||
setTtsProvider: (prefsPath: string, provider: TtsProvider) => void;
|
||||
synthesizeSpeech: (params: TtsRequestParams) => Promise<TtsSynthesisResult>;
|
||||
textToSpeech: (params: TtsRequestParams) => Promise<TtsResult>;
|
||||
textToSpeechTelephony: (params: TtsTelephonyRequestParams) => Promise<TtsTelephonyResult>;
|
||||
};
|
||||
type FacadeModule = TtsRuntimeFacade;
|
||||
|
||||
function loadFacadeModule(): FacadeModule {
|
||||
return loadActivatedBundledPluginPublicSurfaceModuleSync<FacadeModule>({
|
||||
@@ -203,48 +73,15 @@ export const textToSpeech: FacadeModule["textToSpeech"] = createLazyFacadeValue(
|
||||
export const textToSpeechTelephony: FacadeModule["textToSpeechTelephony"] =
|
||||
createLazyFacadeValue("textToSpeechTelephony");
|
||||
|
||||
export type { ResolvedTtsConfig, ResolvedTtsModelOverrides };
|
||||
export type { TtsDirectiveOverrides, TtsDirectiveParseResult };
|
||||
|
||||
export type TtsResult = {
|
||||
success: boolean;
|
||||
audioPath?: string;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
voiceCompatible?: boolean;
|
||||
};
|
||||
|
||||
export type TtsSynthesisResult = {
|
||||
success: boolean;
|
||||
audioBuffer?: Buffer;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
voiceCompatible?: boolean;
|
||||
fileExtension?: string;
|
||||
};
|
||||
|
||||
export type TtsTelephonyResult = {
|
||||
success: boolean;
|
||||
audioBuffer?: Buffer;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
sampleRate?: number;
|
||||
};
|
||||
export type {
|
||||
ResolvedTtsConfig,
|
||||
ResolvedTtsModelOverrides,
|
||||
TtsDirectiveOverrides,
|
||||
TtsDirectiveParseResult,
|
||||
TtsResult,
|
||||
TtsSynthesisResult,
|
||||
TtsTelephonyResult,
|
||||
} from "./tts-runtime.types.js";
|
||||
|
||||
function createLazyFacadeValue<K extends keyof FacadeModule>(key: K): FacadeModule[K] {
|
||||
return ((...args: unknown[]) => {
|
||||
|
||||
191
src/plugin-sdk/tts-runtime.types.ts
Normal file
191
src/plugin-sdk/tts-runtime.types.ts
Normal file
@@ -0,0 +1,191 @@
|
||||
import type { ReplyPayload } from "../auto-reply/reply-payload.js";
|
||||
import type { OpenClawConfig } from "../config/types.openclaw.js";
|
||||
import type { TtsAutoMode, TtsProvider } from "../config/types.tts.js";
|
||||
import type {
|
||||
SpeechProviderConfig,
|
||||
SpeechVoiceOption,
|
||||
TtsDirectiveOverrides,
|
||||
TtsDirectiveParseResult,
|
||||
} from "../tts/provider-types.js";
|
||||
import type { ResolvedTtsConfig, ResolvedTtsModelOverrides } from "../tts/tts-types.js";
|
||||
|
||||
export type { ResolvedTtsConfig, ResolvedTtsModelOverrides };
|
||||
export type { TtsDirectiveOverrides, TtsDirectiveParseResult };
|
||||
|
||||
export type TtsAttemptReasonCode =
|
||||
| "success"
|
||||
| "no_provider_registered"
|
||||
| "not_configured"
|
||||
| "unsupported_for_telephony"
|
||||
| "timeout"
|
||||
| "provider_error";
|
||||
|
||||
export type TtsProviderAttempt = {
|
||||
provider: string;
|
||||
outcome: "success" | "skipped" | "failed";
|
||||
reasonCode: TtsAttemptReasonCode;
|
||||
latencyMs?: number;
|
||||
error?: string;
|
||||
};
|
||||
|
||||
export type TtsStatusEntry = {
|
||||
timestamp: number;
|
||||
success: boolean;
|
||||
textLength: number;
|
||||
summarized: boolean;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
latencyMs?: number;
|
||||
error?: string;
|
||||
};
|
||||
|
||||
export type SummarizeResult = {
|
||||
summary: string;
|
||||
latencyMs: number;
|
||||
inputLength: number;
|
||||
outputLength: number;
|
||||
};
|
||||
|
||||
export type ResolveTtsAutoModeParams = {
|
||||
config: ResolvedTtsConfig;
|
||||
prefsPath: string;
|
||||
sessionAuto?: string;
|
||||
};
|
||||
|
||||
export type ResolveExplicitTtsOverridesParams = {
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
provider?: string;
|
||||
modelId?: string;
|
||||
voiceId?: string;
|
||||
};
|
||||
|
||||
export type TtsRequestParams = {
|
||||
text: string;
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
channel?: string;
|
||||
overrides?: TtsDirectiveOverrides;
|
||||
disableFallback?: boolean;
|
||||
};
|
||||
|
||||
export type TtsTelephonyRequestParams = {
|
||||
text: string;
|
||||
cfg: OpenClawConfig;
|
||||
prefsPath?: string;
|
||||
};
|
||||
|
||||
export type ListSpeechVoicesParams = {
|
||||
provider: string;
|
||||
cfg?: OpenClawConfig;
|
||||
config?: ResolvedTtsConfig;
|
||||
apiKey?: string;
|
||||
baseUrl?: string;
|
||||
};
|
||||
|
||||
export type MaybeApplyTtsToPayloadParams = {
|
||||
payload: ReplyPayload;
|
||||
cfg: OpenClawConfig;
|
||||
channel?: string;
|
||||
kind?: "tool" | "block" | "final";
|
||||
inboundAudio?: boolean;
|
||||
ttsAuto?: string;
|
||||
};
|
||||
|
||||
export type TtsTestFacade = {
|
||||
parseTtsDirectives: (...args: unknown[]) => TtsDirectiveParseResult;
|
||||
resolveModelOverridePolicy: (...args: unknown[]) => ResolvedTtsModelOverrides;
|
||||
supportsNativeVoiceNoteTts: (channel: string | undefined) => boolean;
|
||||
summarizeText: (...args: unknown[]) => Promise<SummarizeResult>;
|
||||
getResolvedSpeechProviderConfig: (
|
||||
config: ResolvedTtsConfig,
|
||||
providerId: string,
|
||||
cfg?: OpenClawConfig,
|
||||
) => SpeechProviderConfig;
|
||||
formatTtsProviderError: (provider: TtsProvider, err: unknown) => string;
|
||||
sanitizeTtsErrorForLog: (err: unknown) => string;
|
||||
};
|
||||
|
||||
export type TtsResult = {
|
||||
success: boolean;
|
||||
audioPath?: string;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
voiceCompatible?: boolean;
|
||||
};
|
||||
|
||||
export type TtsSynthesisResult = {
|
||||
success: boolean;
|
||||
audioBuffer?: Buffer;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
voiceCompatible?: boolean;
|
||||
fileExtension?: string;
|
||||
};
|
||||
|
||||
export type TtsTelephonyResult = {
|
||||
success: boolean;
|
||||
audioBuffer?: Buffer;
|
||||
error?: string;
|
||||
latencyMs?: number;
|
||||
provider?: string;
|
||||
fallbackFrom?: string;
|
||||
attemptedProviders?: string[];
|
||||
attempts?: TtsProviderAttempt[];
|
||||
outputFormat?: string;
|
||||
sampleRate?: number;
|
||||
};
|
||||
|
||||
export type TextToSpeech = (params: TtsRequestParams) => Promise<TtsResult>;
|
||||
export type TextToSpeechTelephony = (
|
||||
params: TtsTelephonyRequestParams,
|
||||
) => Promise<TtsTelephonyResult>;
|
||||
export type ListSpeechVoices = (params: ListSpeechVoicesParams) => Promise<SpeechVoiceOption[]>;
|
||||
|
||||
export type TtsRuntimeFacade = {
|
||||
_test: TtsTestFacade;
|
||||
buildTtsSystemPromptHint: (cfg: OpenClawConfig) => string | undefined;
|
||||
getLastTtsAttempt: () => TtsStatusEntry | undefined;
|
||||
getResolvedSpeechProviderConfig: (
|
||||
config: ResolvedTtsConfig,
|
||||
providerId: string,
|
||||
cfg?: OpenClawConfig,
|
||||
) => SpeechProviderConfig;
|
||||
getTtsMaxLength: (prefsPath: string) => number;
|
||||
getTtsProvider: (config: ResolvedTtsConfig, prefsPath: string) => TtsProvider;
|
||||
isSummarizationEnabled: (prefsPath: string) => boolean;
|
||||
isTtsEnabled: (config: ResolvedTtsConfig, prefsPath: string, sessionAuto?: string) => boolean;
|
||||
isTtsProviderConfigured: (
|
||||
config: ResolvedTtsConfig,
|
||||
provider: TtsProvider,
|
||||
cfg?: OpenClawConfig,
|
||||
) => boolean;
|
||||
listSpeechVoices: ListSpeechVoices;
|
||||
maybeApplyTtsToPayload: (params: MaybeApplyTtsToPayloadParams) => Promise<ReplyPayload>;
|
||||
resolveExplicitTtsOverrides: (params: ResolveExplicitTtsOverridesParams) => TtsDirectiveOverrides;
|
||||
resolveTtsAutoMode: (params: ResolveTtsAutoModeParams) => TtsAutoMode;
|
||||
resolveTtsConfig: (cfg: OpenClawConfig) => ResolvedTtsConfig;
|
||||
resolveTtsPrefsPath: (config: ResolvedTtsConfig) => string;
|
||||
resolveTtsProviderOrder: (primary: TtsProvider, cfg?: OpenClawConfig) => TtsProvider[];
|
||||
setLastTtsAttempt: (entry: TtsStatusEntry | undefined) => void;
|
||||
setSummarizationEnabled: (prefsPath: string, enabled: boolean) => void;
|
||||
setTtsAutoMode: (prefsPath: string, mode: TtsAutoMode) => void;
|
||||
setTtsEnabled: (prefsPath: string, enabled: boolean) => void;
|
||||
setTtsMaxLength: (prefsPath: string, maxLength: number) => void;
|
||||
setTtsProvider: (prefsPath: string, provider: TtsProvider) => void;
|
||||
synthesizeSpeech: (params: TtsRequestParams) => Promise<TtsSynthesisResult>;
|
||||
textToSpeech: TextToSpeech;
|
||||
textToSpeechTelephony: TextToSpeechTelephony;
|
||||
};
|
||||
@@ -5,6 +5,11 @@ import type {
|
||||
import type { HeartbeatRunResult } from "../../infra/heartbeat-wake.js";
|
||||
import type { LogLevel } from "../../logging/levels.js";
|
||||
import type { MediaUnderstandingRuntime } from "../../media-understanding/runtime-types.js";
|
||||
import type {
|
||||
ListSpeechVoices,
|
||||
TextToSpeech,
|
||||
TextToSpeechTelephony,
|
||||
} from "../../plugin-sdk/tts-runtime.types.js";
|
||||
|
||||
export type { HeartbeatRunResult };
|
||||
|
||||
@@ -87,9 +92,9 @@ export type PluginRuntimeCore = {
|
||||
resizeToJpeg: typeof import("../../media/image-ops.js").resizeToJpeg;
|
||||
};
|
||||
tts: {
|
||||
textToSpeech: typeof import("../../tts/tts.js").textToSpeech;
|
||||
textToSpeechTelephony: typeof import("../../tts/tts.js").textToSpeechTelephony;
|
||||
listVoices: typeof import("../../tts/tts.js").listSpeechVoices;
|
||||
textToSpeech: TextToSpeech;
|
||||
textToSpeechTelephony: TextToSpeechTelephony;
|
||||
listVoices: ListSpeechVoices;
|
||||
};
|
||||
mediaUnderstanding: {
|
||||
runFile: MediaUnderstandingRuntime["runMediaUnderstandingFile"];
|
||||
|
||||
Reference in New Issue
Block a user