Switch config to JSON

README.md

@@ -1,6 +1,6 @@
 # lel
 
-X11 transcription daemon that records audio and runs Whisper, logging the transcript.
+X11 transcription daemon that records audio, runs Whisper, logs the transcript, and can optionally run AI post-processing before injecting text.
 
 ## Requirements
 
@@ -13,26 +13,37 @@ X11 transcription daemon that records audio and runs Whisper, logging the transc
 ## Build
 
 ```bash
-go build -o leld ./cmd/leld
-go build -o lelctl ./cmd/lelctl
+make build
 ```
 
 ## Config
 
-Create `~/.config/lel/config.toml`:
+Create `~/.config/lel/config.json`:
 
-```toml
-hotkey = "Cmd+m"
-ffmpeg_input = "pulse:default"
-whisper_model = "base"
-whisper_lang = "en"
-whisper_device = "cpu"
-whisper_extra_args = ""
-record_timeout_sec = 120
-whisper_timeout_sec = 300
-segment_sec = 5
-streaming = false
-injection_backend = "clipboard"
+```json
+{
+  "hotkey": "Cmd+m",
+  "ffmpeg_input": "pulse:default",
+  "ffmpeg_path": "",
+  "whisper_model": "base",
+  "whisper_lang": "en",
+  "whisper_device": "cpu",
+  "whisper_extra_args": "",
+  "record_timeout_sec": 120,
+  "whisper_timeout_sec": 300,
+  "segment_sec": 5,
+  "streaming": false,
+  "injection_backend": "clipboard",
+
+  "ai_enabled": true,
+  "ai_provider": "ollama",
+  "ai_model": "llama3.2:3b",
+  "ai_temperature": 0.0,
+  "ai_system_prompt_file": "",
+  "ai_base_url": "http://localhost:11434",
+  "ai_api_key": "",
+  "ai_timeout_sec": 20
+}
 ```
 
 Env overrides:
@@ -41,11 +52,14 @@ Env overrides:
 - `WHISPER_FFMPEG_IN`
 - `WHISPER_STREAM`, `WHISPER_SEGMENT_SEC`, `WHISPER_TIMEOUT_SEC`
 - `LEL_RECORD_TIMEOUT_SEC`, `LEL_HOTKEY`, `LEL_INJECTION_BACKEND`
+- `LEL_FFMPEG_PATH`
+- `LEL_AI_ENABLED`, `LEL_AI_PROVIDER`, `LEL_AI_MODEL`, `LEL_AI_TEMPERATURE`, `LEL_AI_SYSTEM_PROMPT_FILE`
+- `LEL_AI_BASE_URL`, `LEL_AI_API_KEY`, `LEL_AI_TIMEOUT_SEC`
 
 ## Run manually
 
 ```bash
-./leld --config ~/.config/lel/config.toml
+./leld --config ~/.config/lel/config.json
 ```
 
 Disable the tray icon:
@@ -70,11 +84,26 @@ systemctl --user enable --now lel
 - Press it again to stop and transcribe.
 - The transcript is logged to stderr.
 
+Execution flow (single in-flight state machine):
+
+- `recording` -> `transcribing` -> `processing` (optional) -> `outputting` -> `idle`
+
 Injection backends:
 
 - `clipboard`: copy to clipboard and inject via Ctrl+V (requires `xclip` + `xdotool`)
 - `injection`: type the text with simulated keypresses (requires `xdotool`)
 
+AI providers:
+
+- `ollama`: calls the local Ollama HTTP API (`/api/generate`)
+- `openai_compat`: calls a chat-completions compatible API (`/v1/chat/completions`)
+
+Dependency checks:
+
+- Recording requires `ffmpeg` (or set `ffmpeg_path`)
+- Transcribing uses the `whisper` CLI
+- Outputting requires `xclip` (and `xdotool` for injection backends)
+
 Control:
 
 ```bash