ShortLoop – Replay Audio Calls and Test Your VoiceAI
VoiceAI builders can replay production calls (or) create voice test data set and replay those calls to your VoiceAI. With this(https://app.shortloop.dev/) you can now test end to end flows frequently and in an automated way. We have been tinkering with LLMs and building voice applications for the past few months. As we were trying to improve our voice bots performance (interruptions, latency of tts, llm etc) it became very tedious to test our bot over and over again speaking to it. With real users we would observe edge cases that we didn’t handle before. After making changes…
In plain words
ShortLoop allows VoiceAI builders to replay production call recordings or create custom voice test datasets and replay them against their voice applications. This enables automated, frequent testing of end-to-end flows without manual repetition. The tool helps developers identify and fix edge cases in voice bot performance, such as handling interruptions, latency issues, and varied speech patterns, by replaying the exact same audio scenarios multiple times to test different bot changes.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
VoiceAI builders can replay production calls (or) create voice test data set and replay those calls to your VoiceAI. With this(https://app.shortloop.dev/) you can now test end to end flows frequently and in an automated way. We have been tinkering with LLMs and building voice applications for the past few months. As we were trying to improve our voice bots performance (interruptions, latency of tts, llm etc) it became very tedious to test our bot over and over again speaking to it. With real users we would observe edge cases that we didn’t handle before. After making changes it was difficult to recreate that exact scenario (accent, noise, slow speech or gaps between sentences) and test changes. So we built a way to use any existing call recording and replay ‘user’ portion of the call to the bot. Since LLMs won’t repeat the exact text or follow same flow on each run, it became important to handle those deviations. So took sometime experimenting and came up with this approach: 1. Segment call in to audio snippets of dialogue 2. Create a transcript and understand flow, intent, details of the call. 3. In response to LLM, identify correct user audio snippet to replay based on the flow/context (lot of work here :D) 4. In cases where it is a new question or reconfirmation by bot, generate new text+audio. I think this is pretty cool and would love for you to try it over your VoiceAI bots and give feedback to improve it. Earlier we also built a visualisation for analysing audio calls (player, transcript and basic errors) and included it. Here is a quick demo of how both look: https://youtu.be/j3kRhSxD5P0 What’s next? Thinking of creating more voice based automated tests. Open to ideas and would love to know how you improve your VoiceBots.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2024
the whole month →
- IY
Life & fun · 2024 · ytch.xyz
- IA
Hey there HN! We’re Joe and Stopa, and today we’re open sourcing InstantDB, a client-side database that makes it easy to build real-time and collaborative apps like Notion and Figma. Building modern apps these days involves a lot of schleps. For a basic CRUD app you need to spin up servers, wire up endpoints, integrate auth, add permissions, and then marshal data from the backend to the frontend and back again. If you want to deliver a buttery smooth user experience, you’ll need to add optimistic updates and rollbacks. We do these steps over and over for every feature we build, which can…
Dev tools · 2024 · github.com