← All projects 
Computer Vision
EchoSight
Image captioning that speaks — attention, GenAI, and real TTS
InceptionV3 → Bahdanau attention → GRU captioning trained on Flickr8k, with an LLM expanding each caption and real text-to-speech reading it aloud. The real training run surfaced and fixed two bugs no code review had caught, and a genuine evaluation-split data leak — all honestly re-reported. v0.2.0.
— Stars
— Open Issues
— Language
— License
Commit Activity — past year
GitHub is computing this repo's commit history — usually ready in a few seconds…
Language Breakdown
Recent activity
Live GitHub stats unavailable right now — see the repo directly.
How it was built
EchoSight, from first question to shipped.
StatusShippedall stages complete
Stages4 / 4frame to ship
Headlinev0.2.0current release
Frame Passed
- Image captioning that speaks
- Trained on Flickr8k
Build Passed
- InceptionV3, then Bahdanau attention, then a GRU
- An LLM expands each caption
- Real text-to-speech reads it aloud
Validate Passed
- The real training run surfaced and fixed two bugs no code review had caught
- Found and fixed an evaluation-split data leak, honestly re-reported
Ship Passed
- v0.2.0