Spaces:
No application file
No application file
created readme
Browse files
README.md
CHANGED
|
@@ -1,14 +1,70 @@
|
|
| 1 |
---
|
| 2 |
title: Echo Ledger
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: green
|
| 5 |
-
colorTo:
|
| 6 |
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
-
python_version: '3.13'
|
| 9 |
app_file: app.py
|
| 10 |
pinned: false
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: Echo Ledger
|
| 3 |
+
emoji: 馃帣
|
| 4 |
colorFrom: green
|
| 5 |
+
colorTo: green
|
| 6 |
sdk: gradio
|
| 7 |
+
sdk_version: "5.0.0"
|
|
|
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
+
tags:
|
| 11 |
+
- hackathon
|
| 12 |
+
- build-small-hackathon
|
| 13 |
+
- backyard-ai
|
| 14 |
+
- best-use-of-modal
|
| 15 |
+
- best-demo
|
| 16 |
+
- best-agent
|
| 17 |
+
license: mit
|
| 18 |
---
|
| 19 |
|
| 20 |
+
# Echo Ledger
|
| 21 |
+
|
| 22 |
+
Talk. Get stuff done.
|
| 23 |
+
|
| 24 |
+
You say "email Sarthak about the logs, book a team sync for Thursday, draft an investor outline" -- Echo Ledger hears it, pulls out every task, and hands you ready-to-use drafts. No typing. No forgetting. No $0.02 per token.
|
| 25 |
+
|
| 26 |
+
[Demo video](#) 路 [Social post](#)
|
| 27 |
+
|
| 28 |
+
---
|
| 29 |
+
|
| 30 |
+
## What it does
|
| 31 |
+
|
| 32 |
+
Record a voice memo. Whisper transcribes it. Gemma 4 12B reads the transcript and lists every task you mentioned. You pick which ones to automate. It generates the draft -- email body, calendar invite, presentation outline -- and gives you a one-tap link to send or download it. Everything lands in an activity log you can scroll through later.
|
| 33 |
+
|
| 34 |
+
That is the whole thing.
|
| 35 |
+
|
| 36 |
+
---
|
| 37 |
+
|
| 38 |
+
## How to run it
|
| 39 |
+
|
| 40 |
+
```bash
|
| 41 |
+
git clone https://github.com/davezaxh/echo-ledger
|
| 42 |
+
cd echo-ledger
|
| 43 |
+
python -m venv .venv && source .venv/bin/activate
|
| 44 |
+
pip install -r requirements.txt
|
| 45 |
+
cp .env.example .env # add your HF_TOKEN
|
| 46 |
+
python mcp_server.py # terminal 1
|
| 47 |
+
python app.py # terminal 2 -- open localhost:7860
|
| 48 |
+
```
|
| 49 |
+
|
| 50 |
+
Needs Apple Silicon for local inference. Swap `USE_MODAL=true` in `.env` to run on Modal instead.
|
| 51 |
+
|
| 52 |
+
---
|
| 53 |
+
|
| 54 |
+
## Models
|
| 55 |
+
|
| 56 |
+
| Model | Params | Role |
|
| 57 |
+
|---|---|---|
|
| 58 |
+
| Whisper medium | 769M | Transcription |
|
| 59 |
+
| Gemma 4 12B | 11.95B | Extraction + drafting |
|
| 60 |
+
| Total | ~12.8B | Under the 32B cap |
|
| 61 |
+
|
| 62 |
+
---
|
| 63 |
+
|
| 64 |
+
## Built with
|
| 65 |
+
|
| 66 |
+
Gradio 5 -- faster-whisper -- mlx-lm -- Modal A10G -- SQLite -- FastAPI
|
| 67 |
+
|
| 68 |
+
---
|
| 69 |
+
|
| 70 |
+
*Built for the Hugging Face Build Small Hackathon. Runs on a Mac Mini M4. No subscriptions harmed in the making of this app.*
|