syntaxbullet/calliope
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

chore: update readme.md

Browse Source
This commit is contained in:
syntaxbullet
2026-02-18 16:18:16 +01:00
parent 134200d345
commit 4441f4046f
1 changed files with 24 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
README.md
View File

@@ -8,7 +8,8 @@ Calliope lives in your menu bar and turns speech into text in any application. P
- **Menu bar native** — Runs quietly in the macOS menu bar, always one hotkey away - **Menu bar native** — Runs quietly in the macOS menu bar, always one hotkey away
- **Universal text input** — Types transcribed text directly into any focused application via Quartz events or clipboard paste - **Universal text input** — Types transcribed text directly into any focused application via Quartz events or clipboard paste
- **On-device transcription** — Powered by OpenAI Whisper models via Hugging Face Transformers, accelerated with MPS on Apple Silicon - **On-device transcription** — Powered by OpenAI Whisper models via `mlx-whisper`, natively accelerated on Apple Silicon with no MPS/PyTorch overhead
- **Auto-stop on silence** — Recording stops automatically after a configurable period of silence, so you don't have to press the hotkey again
- **LLM post-processing** — Optional grammar and punctuation correction using local MLX language models - **LLM post-processing** — Optional grammar and punctuation correction using local MLX language models
- **Live waveform overlay** — Floating visual feedback showing audio levels during recording and a pulsing indicator during transcription - **Live waveform overlay** — Floating visual feedback showing audio levels during recording and a pulsing indicator during transcription
- **Dual hotkey modes** — Push-to-talk (hold to record) and toggle (tap to start/stop), both fully configurable - **Dual hotkey modes** — Push-to-talk (hold to record) and toggle (tap to start/stop), both fully configurable
@@ -38,7 +39,7 @@ pip install -e .
calliope # Launch (runs setup wizard on first run) calliope # Launch (runs setup wizard on first run)
calliope setup # Re-run the setup wizard calliope setup # Re-run the setup wizard
calliope --debug # Launch with verbose logging calliope --debug # Launch with verbose logging
calliope --device 2 --model openai/whisper-large-v3 # Override config for this session calliope --device 2 --model mlx-community/whisper-large-v3 # Override config for this session
calliope --version # Print version calliope --version # Print version
``` ```
@@ -56,22 +57,24 @@ Hotkeys are fully configurable through the setup wizard or by editing the config
All settings are stored at `~/.config/calliope/config.yaml`: All settings are stored at `~/.config/calliope/config.yaml`:
```yaml ```yaml
device: null # Microphone index (null = system default) device: null # Microphone index (null = system default)
model: distil-whisper/distil-large-v3 model: mlx-community/whisper-large-v3-turbo
language: auto # Language code or "auto" for detection language: auto # Language code or "auto" for detection
hotkeys: hotkeys:
ptt: ctrl+shift ptt: ctrl+shift
toggle: ctrl+space toggle: ctrl+space
context: "" # Domain-specific terms to improve accuracy context: "" # Domain-specific terms to improve accuracy
typing_mode: char # "char" (keystroke simulation) or "clipboard" (Cmd+V paste) typing_mode: char # "char" (keystroke simulation) or "clipboard" (Cmd+V paste)
typing_delay: 0.005 # Seconds between keystrokes in char mode typing_delay: 0.005 # Seconds between keystrokes in char mode
max_recording_seconds: 300 # Maximum recording duration max_recording_seconds: 300 # Maximum recording duration
silence_threshold: 0.005 # RMS energy below which audio is considered silence silence_threshold: 0.005 # RMS energy below which audio is considered silence
notifications: true # macOS notification banners auto_stop_silence: true # Automatically stop recording after sustained silence
silence_timeout_seconds: 1.5 # Seconds of silence before auto-stop triggers
notifications: true # macOS notification banners
postprocessing: postprocessing:
enabled: false # LLM grammar/punctuation correction enabled: false # LLM grammar/punctuation correction
model: null # Active MLX model model: null # Active MLX model
system_prompt: "..." # Custom post-processing instructions system_prompt: "..." # Custom post-processing instructions
debug: false debug: false
``` ```
@@ -79,13 +82,15 @@ CLI flags override config values for that session.
## Available Models ## Available Models
All models are sourced from Hugging Face and run natively via `mlx-whisper` on Apple Silicon.
| Model | Size | Speed | Accuracy | | Model | Size | Speed | Accuracy |
|-------|------|-------|----------| |-------|------|-------|----------|
| `openai/whisper-base` | ~150 MB | Fastest | Basic | | `mlx-community/whisper-base` | ~150 MB | Fastest | Basic |
| `openai/whisper-small` | ~500 MB | Fast | Good | | `mlx-community/whisper-small` | ~500 MB | Fast | Good |
| `openai/whisper-medium` | ~1.5 GB | Moderate | Better | | `mlx-community/whisper-medium` | ~1.5 GB | Moderate | Better |
| `distil-whisper/distil-large-v3` | ~1.5 GB | Fast | High (default) | | `mlx-community/whisper-large-v3-turbo` | ~1.6 GB | Fast | High (default) |
| `openai/whisper-large-v3` | ~3 GB | Slower | Highest | | `mlx-community/whisper-large-v3` | ~3 GB | Slower | Highest |
## Troubleshooting ## Troubleshooting