Modern NewsTopAskShowBestNew

Show

    Show HN: An API for word timestamps from speech, 75% within 20ms

    by jasperstory · 2 days ago

    Hey HN, we're Yarn (W24) and we make a video editing app for startups, and we built Quebec (https://quebec.yarn.so), an API that gives you accurate word timestamps from speech. We needed it because cutting a word even 50ms off from its true timing can ruin talking head footage (examples in the write up).

    Accurate word timestamps usually come from a forced aligner, which takes the speech audio and a transcript and works out exactly when each word starts and ends. We weren't satisfied with the accuracy of Montreal Forced Aligner (MFA) or new models like ElevenLabs Cue (https://elevenlabs.io/blog/cue-aligner). Giving LLMs access to a waveform/spectrogram didn't help much either.

    So we built our own. There are two particularly interesting parts: - We wrote our own decoder around MFA's acoustic model that only lets each word land within 200ms of the rough timings. That made search small enough to solve exactly, so we don't need to use MFA's beam search. - We added a small CNN, built on the first two layers of Whisper Tiny, that nudges each word boundary by up to 50ms. We couldn't find enough hand-labelled speech that allowed for commercial use, so we had Gemini label our own (17k word boundaries). It took a couple of hours and cost under $200. A net trained on Gemini's labels scored the same as one trained on human labels (73.9% vs. 74%).

    One way to measure accuracy is % of word boundaries within 20ms of hand labels. Here are the results on Buckeye (hand-labelled American speech that we didn't train on):

    - Quebec 75%

    - MFA 61%

    - ElevenLabs Cue 30%

    If you want to try it, it's free until tomorrow night. You can point your agent at our docs at https://quebec.yarn.so, and send audio on its own or with your own transcript. After that, it costs $0.32 per hour with no signup (minimum $10 top-up). We serve it at cost. Caveats: only tested in English and slower than ElevenLabs.

    Let us know if Quebec works well for you, and send us any samples that work badly (you can email support at yarn dot so). Also let us know if you've built or found a better aligner. In theory we would just switch to a better aligner, although we would probably just push a bit harder to make ours better.

    Thanks!

    Jasper

    22|quebec.yarn.so|3 comments

    Show HN: Mastra Factory, the open-source software factory we use to build Mastra

    by codekarate · about 17 hours ago

    Hi HN, we're Shane, Sam, and Abhi.

    Almost two years ago, we shared Mastra, our open-source TypeScript agent framework, on Show HN: https://news.ycombinator.com/item?id=43103073. It got a few hundred upvotes; a couple of weeks later our downloads had doubled, and we were off to the races.

    Today we're sharing Mastra Factory, the open-source software factory we built to help maintain Mastra itself.

    At Mastra, we now average nearly 4M downloads per week, and with Factory our team now merges hundreds of PRs a week across the mastra-ai GitHub org. Last week, Factory opened and merged 238 PRs, alongside 630 merged PRs from humans.

    There's a lot of debate about what a software factory is. For us Factory is a configurable agentic kanban board that's @-taggable in Slack.

    Under the hood, it uses Mastra’s durable agents for long-running sessions, our conversational memory system, a markdown based wiki with source citations to incorporate company knowledge, signals that subscribe to PR comments wired up into sessions, and a customizable cloud sandbox with our coding harness (working on supporting Claude Code, Codex etc).

    When we started out, we modeled Factory around our existing workflow-oriented process (triage, plan, build, review, shipped). Three months later, we find ourselves mostly driving it through Slack vs a web UI.

    We’ve seen more senior engineers kick off sessions in parallel (“triage the 20 oldest issues and let’s evaluate…”). We’re using bug and vuln scanners now, and we’ve closed our repo to outside contributors, asking folks to write detailed issue repros instead.

    You can get started with `npm create factory`.

    It’s fully open-source, so you can run it entirely on your infra (we also offer a hosted version). Comment here, we would love to know what you think!

    https://factory.mastra.ai

    https://github.com/mastra-ai/template-factory

    93||33 comments

    Show HN: ExitPath – Open-source ways out of Slack, Notion and Microsoft 365

    by zerocool0101 · about 18 hours ago

    213|exitpath-ruddy.vercel.app|68 comments

    Show HN: Take Voyager 1 apart, down to a single memory chip

    by raj-shekhar · about 17 hours ago

    Voyager 1 reached one light-day from Earth today.

    On the site you can:

    - see where both Voyagers are right now

    - open a nuclear battery to its 24 plutonium spheres

    - go into the data computer, down to a single memory chip

    - see how the Golden Record's cover explains how to play it

    - watch how Voyager 1 was fixed by radio

    - see which dish on Earth is listening to them, live

    I also made a 4-minute film from it: https://www.youtube.com/watch?v=kpzsRo-c1_A

    195|rshekhar.com|49 comments

    Show HN: Open-source, real-time object tracking for live video feeds

    by aryaman_darda · 3 days ago

    I type "forklift" on a live warehouse camera, and tamserve finds and tracks both forklifts, including the one driving in. Then I add "person" without stopping it.

    tamserve runs Meta's SAM 3 on live camera feeds. SAM 3's own tracker handles one recorded video at a time, runs at about 8 fps, and runs out of memory on long videos. tamserve uses SAM 3 to find objects every few frames and a lightweight tracker (EdgeTAM) to follow them in between, with one copy of each model shared across all your cameras.

    *On one RTX 4090:*

    * *Any object you name:* 4x faster than SAM 3's tracker, at 92% of its accuracy * *People:* 16 cameras at once, 14x less GPU per frame, and slightly more accurate than SAM 3 (89.6 vs 87.8 HOTA) * *Memory:* flat at 3.5 GB on a 3,000 frame video, where SAM 3 runs out at frame 423

    *Worth a look if you're building:*

    * Warehouse or factory safety (forklifts, pallets, hard hats) * Security or retail camera analytics * Home or elder care monitoring * Robotics, wildlife or lab monitoring * Annotation tools for long videos

    `pip install "tamserve[sam3]"`

    Please give me any and all feedback or edit tamserve to meet your needs, and a star helps if you find it useful!

    91|github.com|8 comments

    Show HN: Zpup – Zdog style 3D scenes in JSON, SVG from any angle

    by select1dev · about 4 hours ago

    Hi. I made small project called Zpup. It takes small 3D scene written in JSON and gives back a flat SVG from any angle. There is no JavaScript in the file, so it works in img-tag or README.

    I needed 3D for small project at my company, and Zdog make big impression on me. But Zdog has not had a release for several years, so I wanted to make modern version.

    One thing works different. Zdog sorts each shape by one point. so small thing on big floor gets covered when you turn it. Zpup compares the shapes that overlap. You can see both in the README.

    I wrote it with Claude Code Fable 5.2.

    No canvas renderer, no text shape, and it is 0.2 version.

    Playground: https://rxolve.github.io/zpup

    Feedback is always welcome :)

    5|github.com|0 comments

    Show HN: Type Fighter 2

    by torumitch · about 22 hours ago

    Type Fighter II is an arcade inspired game for people learning to touch type.

    Free, no account, and progress stays local. You can try on mobile, best with a physical keyboard.

    134|typefighter2.com|59 comments

    Show HN: ADE Index – a sourced and ranked index of 975 agentic coding tools

    by jonnyasmar · about 8 hours ago

    17|adeindex.com|3 comments

    Show HN: Robotics needs better CAD, so I wrote a B-rep kernel and 3,487 tests

    by atomarch · about 13 hours ago

    Hey HN, I’m Tony. I’m a co-founder of Springcraft, where we have been doing a lot of CAD modeling for robot parts. Our team kept running into limitations with existing kernels, and so I decided to write a new B-rep kernel and a new comprehensive, kernel-agnostic B-rep conformance test suite. Both are available today under MIT licenses.

    The kernel, springrep, is written in safe Rust with no dependencies. It covers a broad range of our mechanical CAD needs, including advanced fillets, sweeps, lofts, shells, offsets, STEP and exact measurements and queries. Claude Code made it feasible to write this, but it took a lot of time and care to get it right.

    The test suite, brep-conformance, has 3,487 cases covering the modeling operations common across kernels: booleans, fillets, chamfers, shells, offsets, sweeps, lofts, STEP, and queries with expected answers coming from closed forms and exact geometry. It caught hundreds of bugs, edge cases, and hot spots in springrep that our own tests had missed, and now springrep passes all of it. We’ve also run it on a few popular open-source kernels and anyone can run it on theirs. This is the first version, so there are surely mistakes in the test and adapters. Corrections, pull requests, and new kernels are welcome.

    kernel: https://github.com/springcraft-ai/springrep conformance suite: https://springcraft-ai.github.io/brep-conformance/

    31|springcraft-ai.github.io|6 comments

    Show HN: Snowchive – Snowden archive of all published documents

    by libroot · about 4 hours ago

    4|snowchive.org|0 comments