Built something I think some of you may find useful: WhisperWeb

It’s a web app for turning audio into text quickly and simply in the browser. Great for voice notes, interviews, rough transcripts, and idea capture.

You can check it out here: https://whisperweb.app

I’d love to hear what you think.
#SpeechToText #Transcription #IndieWeb #WebApp #Productivity

Whisper Web — In‑Browser Speech‑to‑Text

Transcribe audio privately in your browser. No uploads. Try it now at whisperweb.app.

Whisper Web

🔊 #Newsletter alert!

#Transcription. #Translation. #Dubbing. #VoiceOvers.

#AI is doing (almost) all of it now. 🤖

*Babylon* – curated by our colleague Mirko Lorenz – explores and explains what that actually means.

For issue No. 1 & free subscription, follow this link:

https://www.babylon-newsletter.com/p/issue-1

Babylon Newsletter - Issue #1

Exploring the evolution of AI language technologies and the future of content.What are new developments? What works?

Babylon - Language Technology Newsletter

My injured finger is still too sore to do any typing. Luckily, #VoiceInk, an #opensource dictation app has been coming in clutch for Mac input needs while my finger heals.

It does the audio #transcription/processing locally (which is the only way I think voice data should be managed). I tried it out with the free trial and it's been surprisingly accurate and easy to use.

I recommend it if you're into dictation apps and I'd trust this more than Apple's built-in dication.

#dictation #privacy

What is the evolutionary significance of alternative #transcription initiation, #splicing & #polyadenylation? This study of 75 metazoan species suggests that most transcript diversity reflects deleterious RNA processing errors rather than adaptive function @PLOSBiology https://plos.io/3Pcs2e6
Transcript diversity reflects deleterious RNA processing errors shaped by population size in metazoans

Alternative transcription initiation, splicing and polyadenylation generate extensive transcript diversity in eukaryotes, but its evolutionary significance has been disputed. This study analyses 166 transcriptomes across 75 metazoan species to show that transcript diversity generally decreases with effective population size, supporting the view that most transcript diversity reflects deleterious RNA processing errors rather than adaptive functions.

Here are some of the ones we've heard of; the options are a bit overwhelming. If you've used any of these (or others) and could share your thoughts on use cases or limitations, that would be so helpful!

https://amical.ai/
https://ethic8.com/ - Canadian but not open source
https://meetily.ai
https://openwhispr.com/
https://epicenter.so/whispering/

#opensource #transcription #privacy #DigitalSecurity #DataSovereignty #AI

Amical - Open source AI Dictation and Note-taking

Type 10x faster, no keyboards needed. Fast, Accurate, Context-aware and Private.

What's everyone's favourite meeting transcription software that's not Otter.ai?

I'd love to hear your recommendations for specific open-source data-secure transcription software for video-conference (and in-person) meeting notes! It needs to be suitable for grassroots orgs, particularly folks who prefer not to self-host.

#opensource #transcription #privacy #DigitalSecurity #DataSovereignty #AI

#JeudiAutoEdition Retrouvez ma #transcription pour #trio à #cordes du Clavier bien tempéré de Jean-Sébastien #Bach ➡️ https://nicolashussein.fr/produit/jean-sebastien-bach-le-clavier-bien-tempere-extraits/

#violon #alto #violoncelle #préludes #fugues #musique #musiqueClassique #classicalMusic #music #myWork #partition

Disponible sur Amazon, The Book Edition ou Planète Partitions 😍🎶🎻

Just ran Whisper (OpenAI) completely locally on my system (RX 6700 XT / 16 GB RAM).

Whisper is an open source speech recognition model that can transcribe audio, generate subtitles, and even translate between languages.

Test video: The Reason Why Cancer is so Hard to Beat by Kurzgesagt - In a Nutshell
(https://www.youtube.com/watch?v=uoJwt9l-XhQ)

Setup:

- Whisper installed via pip
- Model: small (fast, good enough for English)
- GPU acceleration via ROCm

Result:
~98% accurate transcription with only a few minor errors, already solid for generating subtitles.

Next steps / possibilities:

- Auto-generate subtitles (.srt)
- Correct subtitles with a local LLM
- Translate speech
- Burn subtitles directly into videos

Video workflow:

- Recorded with OBS
- Edited in Kdenlive
- Transcoded with VAAPI (H.264)

No cloud, real hardware.
Everything runs on Linux, so anyone can set this up.
No GPU? No problem, you can also run it using PyTorch’s CPU backend, just much slower.

Background music: End of Me - Ashes Remain [Female Rock Cover by Kryx] (https://www.youtube.com/watch?v=E430M8lKim8)


#Whisper #OpenAI #ROCm #AMD #Linux #SpeechToText #Transcription #Subtitles #FOSS #OpenSource #OfflineAI #localai #Fediverse #nocloud

Library of Congress: Celebrating Seven Years of By the People. “Happy Spring to all our By the People crowdsourced transcription program volunteers! Every year, the By the People team publishes a “happy birthday to us” blog post right here on the Signal (you can check out previous years’ editions here and here). We turned seven during the Fall of last year and we’ve been waiting to […]

https://rbfirehose.com/2026/03/17/library-of-congress-celebrating-seven-years-of-by-the-people/
Library of Congress: Celebrating Seven Years of By the People

Library of Congress: Celebrating Seven Years of By the People. “Happy Spring to all our By the People crowdsourced transcription program volunteers! Every year, the By the People team publish…

ResearchBuzz: Firehose

TestingCatalog News (@testingcatalog)

Hypescribe로 보이는 서비스가 YouTube, TikTok, Instagram, Zoom 통화, Google Meet, 음성 메모, MP4 등 다양한 소스의 음성/비디오를 지원하며 100개 이상의 언어를 처리하고 최대 99% 정확도를 주장합니다. 파일 길이 제한 없이 토큰 기반 과금 모델을 적용하고 있으며, hypescribe.com에서 무료 테스트가 가능하다고 안내하고 있습니다.

https://x.com/testingcatalog/status/2033567051466371556

#transcription #speechtotext #hypescribe #multilingual

TestingCatalog News 🗞 (@testingcatalog) on X

It reportedly works with YouTube, TikTok, Instagram, Zoom calls, Google Meet, voice memos, MP4s, and more. 100+ languages. Up to 99% accuracy. Token-based billing with no file length limits. Free to test it on https://t.co/P2McvyuKBC 👇

X (formerly Twitter)