Sherpa Voice vs LLM Bench

Sherpa Voice is an open-source demo app that showcases what on-device speech and audio AI can do — directly on your phone, with no cloud and no internet required. Whether you are a developer evaluating libraries for your next app, a researcher exploring mobile inference, or just curious about offline AI, Sherpa Voice lets you try 10 speech and audio features hands-on. WHAT YOU CAN TRY - Speech Recognition: Transcribe speech to text from files or live microphone input. Multiple languages supported. - Text-to-Speech: Generate natural-sounding speech from text with various voices and speakers. - Voice Activity Detection: Detect speech vs silence in real-time audio. - Keyword Spotting: Listen for custom keywords without full transcription. - Speaker Identification: Recognize speakers by their voice. - Speaker Diarization: Figure out who spoke when in a conversation. - Audio Tagging: Classify sounds — music, speech, environmental noise. - Language Identification: Detect which language is being spoken. - Speech Enhancement: Remove background noise from recordings. - Punctuation: Add punctuation to raw transcripts. Each feature comes with a downloadable model catalog so you can compare different models for speed, size, and accuracy. FOR DEVELOPERS Sherpa Voice is built with React Native and Expo using two open-source libraries: - sherpa-onnx: C++ inference engine for speech models, with mobile-optimized ONNX Runtime - expo-audio-stream: Real-time audio recording, analysis, and visualization for React Native Everything you see in this app can be integrated into your own project. The full source code is available on GitHub — use it as a reference, fork it, or contribute. HOW IT WORKS 1. Pick a feature (Speech Recognition, TTS, etc.) 2. Download a model from the built-in catalog (one-time, typically 15-100 MB) 3. Use the feature — all processing runs locally on your device PRIVACY No accounts. No tracking. No data collection. Audio never leaves your device.

LLM Bench is an iOS test bench for developers comparing on-device LLM behavior under real device constraints. Run the same prompt against selected local model targets and compare the evidence before choosing what belongs in your app. LLM Bench focuses on practical signals: load time, first output time, generation time, output speed, memory, thermal state, model readiness, raw output text, failures, and saved run history. Use LLM Bench to: - Compare GPU-capable local model runs on iPhone and iPad. - Run repeatable prompts against selected model targets. - Inspect model output, timing, memory, and thermal state in one place. - Save benchmark runs for later review. - Export benchmark evidence as CSV. - Add models from Hugging Face URLs, direct file URLs, or local imports when the artifact format is supported. LLM Bench is built for technical builders who want a compact measurement tool, not a leaderboard or generic AI chat app. It helps you answer the practical question: which local model behaves well enough on this device for this use case? The free tier includes one completed benchmark run. Subscribing unlocks ongoing benchmarking, saved comparison work, and export workflows. Premium is available as an auto-renewing subscription (monthly or yearly), or a one-time Lifetime purchase. Payment is charged to your Apple ID at purchase; subscriptions renew automatically unless canceled at least 24 hours before the period ends. Manage or cancel anytime in your App Store account settings. Privacy Policy: https://thedandano.github.io/michilotl-studios-docs/apps/llm-bench/privacy/ Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/