feat(voice): voice/speech integration with TTS, STT, and realtime providers

.env.example

@@ -405,16 +405,6 @@ NEUROLINK_IMAGE_MAX_SIZE=10485760
 # NEUROLINK_IMAGE_CACHE_TTL_MS=3600000          # 1 hour TTL (longer retention)
 # NEUROLINK_IMAGE_MAX_SIZE=5242880              # 5MB max per image (smaller limit)
 
-# =============================================================================
-# CONVERSATION TITLE GENERATION (Optional)
-# =============================================================================
-# Customize the prompt used to generate conversation titles from user messages.
-# Use ${userMessage} placeholder to include the user's message in the prompt.
-# If not set, uses default prompt optimized for 20-character titles.
-#
-NEUROLINK_TITLE_PROMPT='Create a short 3-word title for: ${userMessage}'
-#
-
 # =============================================================================
 # CONVERSATION MEMORY CONFIGURATION (Optional)
 # =============================================================================
@@ -806,61 +796,33 @@ PICOVOICE_ACCESS_KEY=
 VOICE_LLM_MODEL=gpt-4o-automatic
 VOICE_LLM_PROVIDER=azure
 
-# =============================================================================
-# DEEPSEEK CONFIGURATION
-# =============================================================================
-# Get an API key at https://platform.deepseek.com/api_keys
-DEEPSEEK_API_KEY=
-# Optional: override default model (deepseek-chat | deepseek-reasoner)
-DEEPSEEK_MODEL=deepseek-chat
-# Optional: override default base URL
-# DEEPSEEK_BASE_URL=https://api.deepseek.com
+# === Voice Provider Credentials (TTS / STT) ===
+# ElevenLabs TTS (https://elevenlabs.io)
+ELEVENLABS_API_KEY=
 
-# =============================================================================
-# NVIDIA NIM CONFIGURATION
-# =============================================================================
-# Get an API key at https://build.nvidia.com/settings/api-keys
-NVIDIA_NIM_API_KEY=
-# Optional: override default model (browse https://build.nvidia.com/models)
-NVIDIA_NIM_MODEL=meta/llama-3.3-70b-instruct
-# Optional: override default base URL (use for self-hosted NIM)
-# NVIDIA_NIM_BASE_URL=https://integrate.api.nvidia.com/v1
-# Optional NIM extras (rarely needed, leave commented)
-# NVIDIA_NIM_TOP_K=
-# NVIDIA_NIM_MIN_P=
-# NVIDIA_NIM_REPETITION_PENALTY=
-# NVIDIA_NIM_MIN_TOKENS=
-# NVIDIA_NIM_CHAT_TEMPLATE=
+# Deepgram STT (https://deepgram.com)
+DEEPGRAM_API_KEY=
 
-# =============================================================================
-# LM STUDIO CONFIGURATION (local provider)
-# =============================================================================
-# Install LM Studio: https://lmstudio.ai/
-# Load a model in the app and click "Start Server"
-LM_STUDIO_BASE_URL=http://localhost:1234/v1
-# Optional: explicit model id (blank = auto-discover from /v1/models)
-LM_STUDIO_MODEL=
-# Optional: API key (not required for stock LM Studio; use only when running
-# behind an auth-proxying reverse-proxy)
-# LM_STUDIO_API_KEY=
+# Azure Cognitive Services Speech (TTS + STT)
+# Get from: Azure Portal → Cognitive Services → Speech → Keys and Endpoint
+AZURE_SPEECH_KEY=
+AZURE_SPEECH_REGION=eastus
 
-# =============================================================================
-# LLAMA.CPP CONFIGURATION (local provider)
-# =============================================================================
-# Run: ./llama-server -m model.gguf --port 8080 (add --jinja for tool support)
-LLAMACPP_BASE_URL=http://localhost:8080/v1
-# Optional: explicit model id (blank = use whatever model llama-server has loaded)
-LLAMACPP_MODEL=
-# Optional: API key (not required for stock llama-server; use only when running
-# behind an auth-proxying reverse-proxy)
-# LLAMACPP_API_KEY=
+# Google API Key (for Gemini Live realtime voice)
+GOOGLE_API_KEY=
+
+# === Voice Provider Credentials (TTS / STT) ===
+# ElevenLabs TTS (https://elevenlabs.io)
+ELEVENLABS_API_KEY=
+
+# Deepgram STT (https://deepgram.com)
+DEEPGRAM_API_KEY=
+
+# Azure Cognitive Services Speech (TTS + STT)
+# Get from: Azure Portal → Cognitive Services → Speech → Keys and Endpoint
+AZURE_SPEECH_KEY=
+AZURE_SPEECH_REGION=eastus
+
+# Google API Key (for Gemini Live realtime voice)
+GOOGLE_API_KEY=
 
-# =============================================================================
-# TEST-ONLY CREDENTIALS (used by test/continuous-test-suite-credentials.ts and
-# test/continuous-test-suite-new-providers.ts to verify per-call overrides
-# without depending on the runtime env vars above)
-# =============================================================================
-# TEST_DEEPSEEK_API_KEY=
-# TEST_NVIDIA_NIM_API_KEY=
-# TEST_LM_STUDIO_BASE_URL=
-# TEST_LLAMACPP_BASE_URL=

CHANGELOG.md

@@ -1,115 +1,3 @@
-## [9.60.1](https://github.com/juspay/neurolink/compare/v9.60.0...v9.60.1) (2026-04-30)
-
-### Bug Fixes
-
-- **(proxy):** validate pnpm global store compatibility before auto-update install ([ac573ad](https://github.com/juspay/neurolink/commit/ac573adc688aa28376d4b39ffa8e6bb7539cb40e))
-
-## [9.60.0](https://github.com/juspay/neurolink/compare/v9.59.6...v9.60.0) (2026-04-30)
-
-### Features
-
-- **(providers):** integrate DeepSeek, NVIDIA NIM, LM Studio, llama.cpp ([c829f4d](https://github.com/juspay/neurolink/commit/c829f4dea09bf3a6eae08c4902f9293bfb6c05f6))
-
-## [9.59.6](https://github.com/juspay/neurolink/compare/v9.59.5...v9.59.6) (2026-04-30)
-
-### Bug Fixes
-
-- **(tools):** start execution timeout after HITL approval ([1e6d3e0](https://github.com/juspay/neurolink/commit/1e6d3e044a216b67674116830a3917cf1df7171a))
-
-## [9.59.5](https://github.com/juspay/neurolink/compare/v9.59.4...v9.59.5) (2026-04-29)
-
-### Bug Fixes
-
-- **(routing):** dual-mode image text fallback + skip video-frame hijack on structured output ([97b2373](https://github.com/juspay/neurolink/commit/97b2373a793e56414a2ca41009efc72d9a574999))
-
-## [9.59.4](https://github.com/juspay/neurolink/compare/v9.59.3...v9.59.4) (2026-04-27)
-
-### Bug Fixes
-
-- **(proxy):** replace blocking quiet-gate with best-effort wait for auto-updates ([defd6e0](https://github.com/juspay/neurolink/commit/defd6e0f177abea19e490ebe8bd8ea492f6bebca))
-
-## [9.59.3](https://github.com/juspay/neurolink/compare/v9.59.2...v9.59.3) (2026-04-27)
-
-### Bug Fixes
-
-- **(observability):** enrich NoOutputGeneratedError sentinel chunk metadata + actually trigger the catch path the production bug needs ([6854af1](https://github.com/juspay/neurolink/commit/6854af103688dd20093bbfef15e2252b4b502b51))
-
-## [9.59.2](https://github.com/juspay/neurolink/compare/v9.59.1...v9.59.2) (2026-04-26)
-
-### Bug Fixes
-
-- **(context):** pre-dispatch compaction + hard cap for inline conversationMessages on both generate and stream paths + compaction.insufficient event ([d39739f](https://github.com/juspay/neurolink/commit/d39739fc6ac01aa64e309481e0a8fe53525e9c7f))
-
-## [9.59.1](https://github.com/juspay/neurolink/compare/v9.59.0...v9.59.1) (2026-04-26)
-
-### Bug Fixes
-
-- **(observability):** emit generation:end exactly once on stream finalize ([9bd2cd0](https://github.com/juspay/neurolink/commit/9bd2cd0a16484fa93ef8c5aaa018b4323632f094))
-
-## [9.59.0](https://github.com/juspay/neurolink/compare/v9.58.0...v9.59.0) (2026-04-26)
-
-### Features
-
-- **(errors):** typed ModelAccessDeniedError + sdk.checkCredentials() API ([1ffc5bc](https://github.com/juspay/neurolink/commit/1ffc5bc44ce411086f130cc7ee33cb094290b108))
-
-## [9.58.0](https://github.com/juspay/neurolink/compare/v9.57.1...v9.58.0) (2026-04-26)
-
-### Features
-
-- **(fallback):** providerFallback callback + modelChain config for centralized policy ([92e5026](https://github.com/juspay/neurolink/commit/92e5026ac48ca98c640cd1793b4c194c8b84a128))
-
-## [9.57.1](https://github.com/juspay/neurolink/compare/v9.57.0...v9.57.1) (2026-04-25)
-
-### Bug Fixes
-
-- **(conversation-memory):** stop persisting abort sentinel; add typed AbortError + read-time filter (SI-069/SI-071) ([595b355](https://github.com/juspay/neurolink/commit/595b3558d9bc4eeb4121f0504b9f077ef5f01729))
-
-## [9.57.0](https://github.com/juspay/neurolink/compare/v9.56.2...v9.57.0) (2026-04-25)
-
-### Features
-
-- **(dynamic-args):** add dynamic argument resolution with context-aware utilities ([673b2a2](https://github.com/juspay/neurolink/commit/673b2a213f6ac095645c670280ae4a2bb22946b5))
-
-## [9.56.2](https://github.com/juspay/neurolink/compare/v9.56.1...v9.56.2) (2026-04-24)
-
-### Bug Fixes
-
-- **(files):** honor caller-provided mimetype hint for extension-less buffers ([40276cc](https://github.com/juspay/neurolink/commit/40276cc9abad565089b8161a1e7a9c2eb533df1f))
-
-## [9.56.1](https://github.com/juspay/neurolink/compare/v9.56.0...v9.56.1) (2026-04-21)
-
-### Bug Fixes
-
-- **(context):** Add support to filter out empty content chunks ([5f13d91](https://github.com/juspay/neurolink/commit/5f13d919cb5342dce3c2796fa22436ad6aceb318))
-
-## [9.56.0](https://github.com/juspay/neurolink/compare/v9.55.11...v9.56.0) (2026-04-20)
-
-### Features
-
-- **(logs):** add logs in stream function flow ([730efdc](https://github.com/juspay/neurolink/commit/730efdcca0a509480d0e41c2ee1d0ee25f6b9931))
-
-## [9.55.11](https://github.com/juspay/neurolink/compare/v9.55.10...v9.55.11) (2026-04-20)
-
-### Bug Fixes
-
-- **(observability):** close Curator-reported Langfuse telemetry gaps ([42ed72a](https://github.com/juspay/neurolink/commit/42ed72acf59cca32138b4441c1331f4ed7497454))
-
-## [9.55.10](https://github.com/juspay/neurolink/compare/v9.55.9...v9.55.10) (2026-04-19)
-
-## [9.55.9](https://github.com/juspay/neurolink/compare/v9.55.8...v9.55.9) (2026-04-19)
-
-## [9.55.8](https://github.com/juspay/neurolink/compare/v9.55.7...v9.55.8) (2026-04-19)
-
-## [9.55.7](https://github.com/juspay/neurolink/compare/v9.55.6...v9.55.7) (2026-04-19)
-
-## [9.55.6](https://github.com/juspay/neurolink/compare/v9.55.5...v9.55.6) (2026-04-18)
-
-## [9.55.5](https://github.com/juspay/neurolink/compare/v9.55.4...v9.55.5) (2026-04-18)
-
-## [9.55.4](https://github.com/juspay/neurolink/compare/v9.55.3...v9.55.4) (2026-04-18)
-
-## [9.55.3](https://github.com/juspay/neurolink/compare/v9.55.2...v9.55.3) (2026-04-18)
-
 ## [9.55.2](https://github.com/juspay/neurolink/compare/v9.55.1...v9.55.2) (2026-04-18)
 
 ## [9.55.1](https://github.com/juspay/neurolink/compare/v9.55.0...v9.55.1) (2026-04-18)