Desktop app या local gateway से speech workflows चलाएं.Get startedLocal Gateway APICall the API locally through the EdgeSpeak desktop gateway, or run a headless API service for other trusted machines.→JSON OutputRead the structured transcript and model JSON contracts returned by EdgeSpeak desktop exports, CLI commands, and the local gateway.→LanguagesOne language code format across transcription, alignment, text normalization, speech, Realtime, MCP and the CLI, with the languages each capability supports.→TranscriptionTranscription and speakersUpload audio for transcription, speaker timelines, embeddings and similarity, with response examples.→Text processingSegment sentences or paragraphs and normalize written or spoken text, with complete request and response examples.→SpeechSpeech and voice libraryFind compatible voices, generate WAV or speech events, and add or delete reusable voices.→AlignmentForced alignmentAlign known text with audio to get word and segment timestamps, read progress, and handle concurrent requests.→ChatLanguage and visionCall Responses, Chat Completions, Messages and tokenization, and understand streaming and multimodal inputs.→Realtime sessionsRun a complete WebSocket exchange, send audio, and recognize completion and error events.→TranslationTranslation requirementsWrite background info, translation preferences, dictionary translations, and style so EdgeSpeak translation follows them reliably.→ModelsModels and serviceDiscover capabilities, check health, download and load models, with requests and example responses.→IntegrationsCLIInstall the EdgeSpeak CLI and run local transcription, speaker, speech, model, and agent workflows from the terminal.→Python SDKUse the EdgeSpeak Python SDK for local transcription, alignment, and speaker analysis through the CLI.→MCP SetupConnect EdgeSpeak to any MCP client that supports stdio: transport choice, the tool inventory, and worked request examples.→Agent SkillInstall the EdgeSpeak Agent Skills so AI agents can run local transcription, alignment, segmentation, speech, and subtitle workflows on device.→APIAll endpointsEvery HTTP endpoint of the local gateway, by category.→API changelogChanges to the local gateway HTTP API, Realtime events and MCP tools.→
