InPython in Plain EnglishbyZhang Zhaoyu·Oct 13Stop Sending Bad Audio to Speech-to-Text: Build a Python Preflight Gate with FFmpegA reproducible pipeline for measuring audio structure, level, silence, and clipping risk before transcriptionA response icon1A response icon1
Mia Efoxtech·Aug 7Real-Time Transcription Is a Product Decision, Not a Model UpgradeGPT-Transcribe costs $0.0045 per minute. GPT-Live-Transcribe costs $0.017. The extra cost only makes sense when latency changes what the…
AdwumaTech AI·Jul 27Multi-Lingual Ghanaian Speech-to-Text Translation: An IntroductionIntellectual labour has, for a long time, been a bottleneck to productivity. The same could be said for physical labour before the…
Sonu Yadav·Jul 27I Stopped Paying for Speech-to-Text APIs. Whisper.cpp Runs Everything Offline on My Laptop.There are plenty of speech-to-text APIs available today. Google Speech-to-Text. OpenAI Whisper API. Azure Speech. AWS Transcribe. They all…
Mimi Chen·Jul 21Stop DDoSing Yourself: A Three-Layer Client for Rate-Limited TTS APIsHow 10 concurrent requests against a capacity-3 API went from 6 user-facing errors to zero — with ~30 lines of Python.
Itsuki·Jul 15Swift/MacOS OnDevice STT. Which Model To Choose? (MLX)Benchmarks are useless. Here is what happens when running on MY Device. Disclaimer: All are pretty BAD!A response icon1A response icon1
InOpen IntelligencebyDr. Fadi Shaar·Jul 8Whisper Without the Headache: A Complete Walkthrough of the transcribe-anything Command Line…A response icon1A response icon1
Craig Walls·Jun 15Spring AI Recipe: Talking to AIWhy type your prompts when you can just say them out loud? In this recipe, we’ll add voice input to our chat loop by recording audio from…A response icon1A response icon1
InLevel Up CodingbyItsuki·May 28Swift/iOS: A Better(?) Way to Make A Dictation AppNo Keyboard extension + Open Main App Only On First Record + No Mic Running ALL TIME + Auto Back / Home Behavior without Private APIs
AlterSquare·May 20Voice Agent Latency: Where the 2–3 Second Delay Actually Lives in the Pipeline and How to Reduce ItTL;DR: The 2–3 second delay in voice agents is not random — it’s five sequential stages stacking up: VAD, STT, LLM, TTS, and network…
