56 projects
production-anomaly
Catch abnormal production rates, downtime and process drift on a manufacturing line
audio-anomaly
Detect unusual sounds in machine or environment recordings without labelled faults
offline-stt-router
Find which speech-to-text engines are installed locally and pick the right one for this machine and language
call-ai-metrics
Measure talk time, interruptions, silence and pace in a two-party call
voice-commands-ai
Turn spoken or transcribed commands into structured Python actions
speaker-diarize-lite
Work out who spoke when in a recording, with a lightweight method or a model you supply
audio-clean-ai
Remove steady background noise from speech recordings, without a model
video-event-detector
Detect falls, crowding, abandoned objects and sudden movement from a sequence of frames
object-counter-ai
Count objects in images or across video frames, with a detector you supply or a simple built-in one
image-dedup-ai
Find duplicate and near-duplicate images across a large folder, fast, with an index you can keep
document-quality
Decide whether a scanned page is good enough to OCR before you spend money OCRing it
ocr-cleaner
Prepare a scanned page so OCR reads it better: deskew, denoise, threshold
vision-anomaly
Spot unusual images against a set of normal ones, without labelled defects
speech-quality
Measure whether a voice recording is clean enough to transcribe or publish
voice-activity-ai
Find where someone is actually speaking in a recording, and where it is just noise
camera-health
Detect an obstructed, defocused, dark, frozen or tampered camera from its own frames
image-redactor
Blur or mask faces, plates and regions you mark, with a detector you supply or a simple built-in one
multilingual-text
Detect language, normalise and transliterate text, with one consistent API
smart-crop-ai
Crop to the interesting part of an image instead of the middle of it
document-memory
Give an application a persistent, searchable memory of documents and conversations
predictive-maintenance
Estimate how close equipment is to failure from sensor history, and train a failure model when you have labels
image-quality-ai
Detect blur, darkness, overexposure, noise, low contrast and bad framing in photos before they reach a model
meeting-intelligence
Turn a meeting transcript into decisions, action items and a summary
llm-router-lite
Send each prompt to the cheapest model that can handle it, and fall back when one fails
rag-quality-check
Measure whether a retrieval system is actually retrieving the right things
hallucination-check
Check an answer against the sources it claims to use and flag every unsupported sentence
semantic-dedup
Remove passages that repeat the same meaning, not just the same words
text-quality-ai
Score text for readability, repetition, structure and clarity, and say what to fix
ml-pipeline-kit
Build a preprocess, predict, validate and log pipeline in a few lines, with every step checked
ml-inference-profiler
Find the slow step in an ML inference pipeline, from preprocessing to postprocessing
model-watchdog
Lightweight production monitoring for any ML model: log predictions, catch drift and silent failure
offline-ml
Detect the machine you are on and pick a model configuration that will actually fit and run
model-benchmark
Benchmark several models on the same task and compare latency, memory and accuracy side by side
sensor-anomaly
Spot abnormal behaviour across many industrial sensor channels at once, including faults only visible between channels
energy-analyzer-ai
Find unusual energy consumption, explain what changed, and estimate what it is costing
quality-predictor
Predict product quality from manufacturing parameters before final inspection, and see which settings drive it
timeseries-anomaly
Find anomalies in any time series or IoT signal with one call, no model training required
rule-auto-label
Generate labels for text or tabular data from rules, then extend them with a lightweight ML model and an optional LLM hook
machine-health
A single continuously updated 0-100 health score per machine, combining many sensors and rules
dataframe-schema-guard
Stop ML pipelines from breaking when incoming data changes shape: infer a DataFrame schema once, then validate or enforce it forever
privacy-scan-ml
Find personal data in datasets before it leaks into models: emails, phones, Aadhaar, PAN, cards, IPs, addresses and more
dataset-splitter
Leakage-safe train/validation/test splits in one call: stratified, grouped, time-aware, and checked
smartclean-df
Automatically detects and fixes missing values, duplicates, outliers, inconsistent formats and dirty columns in tabular data
ml-feature-check
Catch useless, redundant, leaking and suspicious features before you train on them
synthetic-tabular
Generate realistic synthetic tabular data that preserves distributions and correlations, without a GPU
near-dupes
Find near-duplicate text, records and images with one call, then dedupe keeping the best copy
dataset-health
One-call health report for any CSV or Parquet dataset: missingness, imbalance, leakage, anomalies, correlations
data-drift-lite
Detect whether production data has drifted from training data, column by column, with a single call
document-ai-toolkit
Comprehensive document processing toolkit for AI/ML applications
context-window-manager
Production-ready LLM context window optimization and management
bias-fairness-auditor
Production-ready ML fairness auditing with bias detection and mitigation
training-data-debugger
Find and fix issues in your ML training data - duplicates, label errors, outliers, and more
model-drift-detector
Production monitoring for ML model drift - detect data drift, concept drift, and performance degradation
hallucination-detector
Production-ready hallucination detection for LLM outputs
production-rag
Enterprise-ready Retrieval-Augmented Generation framework with superior performance, reliability, and observability
sonytech
None
