#transcribe — Public Fediverse posts

Amazon Transcribe カスタム言語モデルで「お食事券」と「汚職事件」を聞き分ける
https://qiita.com/mksamba/items/6b38ac0d49eb7d060fee?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items

#qiita #AWS #Transcribe #AmazonTranscribe

#RealtimeTranscription for #PyConIT 🎙️

Whisper hallucinated on real conference audio: a problem I skipped for time: everything else to build.

🔮 Spoiler: #AmazonTranscribe doesn't generate text, it decodes it.

In the article: the choices I made, and the stories you only find when plugging things in (Linux audio is a zoo) 😄

https://alessandra.bilardi.net/diary/articles/2026-04/realtime-transcription-choices-and-stories-for-pycon-it.en

#DiaryOfALazyDeveloper #aws #transcribe #docker #fastapi

#RealtimeTranscription for #PyConIT 🎙️

Whisper hallucinated on real conference audio: a problem I skipped for time: everything else to build.

🔮 Spoiler: #AmazonTranscribe doesn't generate text, it decodes it.

In the article: the choices I made, and the stories you only find when plugging things in (Linux audio is a zoo) 😄

https://alessandra.bilardi.net/diary/articles/2026-04/realtime-transcription-choices-and-stories-for-pycon-it.en

#DiaryOfALazyDeveloper #aws #transcribe #docker #fastapi

CreatorCaps detects your video format and suggests the best caption style. Pick your font, highlight color, and preset.

No subscriptions, one-time unlock.

Try it on iPhone & iPad: https://apple.co/4ruCxYI

#transcribe #captions #subtitles #iOS

CreatorCaps detects your video format and suggests the best caption style. Pick your font, highlight color, and preset.

No subscriptions, one-time unlock.

Try it on iPhone & iPad: https://apple.co/4ruCxYI

#transcribe #captions #subtitles #iOS

Cohere Transcribe: Speech Recognition

https://cohere.com/blog/transcribe

#HackerNews #Cohere #Transcribe #Speech #Recognition #AI #Tech #Innovation

Cohere Transcribe: Speech Recognition

https://cohere.com/blog/transcribe

#HackerNews #Cohere #Transcribe #Speech #Recognition #AI #Tech #Innovation

Hey les guitaristes, quand vous repiquez une mélodie sur une tablature, au-delà du côté mécanique case 11, 13 corde de sol, vous le concevez comment votre repiquage ? Vous pensez notes jouées ? Intervales ? Degrés ? Juste numéros de cases ?
J'essaie de réfléchir à comment mettre à profit le temps passé à retranscrire et y mettre du liant.
Merci !
#guitare #musique #retranscription #morceau #repiquage #transcribe

Hey les guitaristes, quand vous repiquez une mélodie sur une tablature, au-delà du côté mécanique case 11, 13 corde de sol, vous le concevez comment votre repiquage ? Vous pensez notes jouées ? Intervales ? Degrés ? Juste numéros de cases ?
J'essaie de réfléchir à comment mettre à profit le temps passé à retranscrire et y mettre du liant.
Merci !
#guitare #musique #retranscription #morceau #repiquage #transcribe

Seriously, people, come on, if you can't be bothered to either #transcribe your screen-snapshotted #image meme-post or put #alt-text on it, maybe you should reconsider whether it's actually important enough to post. I just went past a DOZEN screenshotted-post posts that I'd have boosted if they had alt text or a transcript.

Seriously, people, come on, if you can't be bothered to either #transcribe your screen-snapshotted #image meme-post or put #alt-text on it, maybe you should reconsider whether it's actually important enough to post. I just went past a DOZEN screenshotted-post posts that I'd have boosted if they had alt text or a transcript.

#AITranscription : #LimitlessAI has been acquired by #META #Facebook.

As of December 5, 2025, the service will no longer be available in #Brazil, #China, the #EuropeanUnion, #Israel, #SouthKorea, #Turkey, and the #UnitedKingdom.

Users in these regions will have until December 19, 2025 to download their information, after which, their account and all associated information will be permanently deleted.

#transcription #LiveTranscribeApp #transcribeaudio #transcribeapp #Transcribe

Lucid Voice - Offline Voice to Text for Mac
Offline voice transcription for macOS. Turn speech into perfect text 3x faster with privacy-first AI. No internet required, no subscriptions.
https://www.lucidvoice.app/
#ai #apple #audio #macos #tools #transcribe

Lucid Voice - Offline Voice to Text for Mac
Offline voice transcription for macOS. Turn speech into perfect text 3x faster with privacy-first AI. No internet required, no subscriptions.
https://www.lucidvoice.app/
#ai #apple #audio #macos #tools #transcribe

A Beginners Guide to Extract Text from a Specific Portion of a YouTube Video on Linux

Step-by-Step tutorial: https://ostechnix.com/extract-text-from-specific-youtube-video-part-linux/

#Transcribe #Youtube #Ytdlp #Ffmpeg #Whisper #Commandline #Linux #Script

A Beginners Guide to Extract Text from a Specific Portion of a YouTube Video on Linux

Step-by-Step tutorial: https://ostechnix.com/extract-text-from-specific-youtube-video-part-linux/

#Transcribe #Youtube #Ytdlp #Ffmpeg #Whisper #Commandline #Linux #Script

Transcribe music in abc with syntax highlighting

https://fugue-state.io/app?project=24024aab-22f1-43cc-abef-c1647cc59597

#HackerNews #Transcribe #music #abc #syntax #highlighting #musictech #codingmusic #abcnotation

Transcribe music in abc with syntax highlighting

https://fugue-state.io/app?project=24024aab-22f1-43cc-abef-c1647cc59597

#HackerNews #Transcribe #music #abc #syntax #highlighting #musictech #codingmusic #abcnotation

Two former Harvard students are launching #HaloX, a pair of #AI-powered #smartglasses that #record and #transcribe #conversations, providing real-time information to the wearer. The glasses, priced at £199, aim to enhance intelligence and memory, but privacy concerns arise due to the lack of an external recording indicator. https://techcrunch.com/2025/08/20/harvard-dropouts-to-launch-always-on-ai-smart-glasses-that-listen-and-record-every-conversation/?eicker.news #tech #media #news

Two former Harvard students are launching #HaloX, a pair of #AI-powered #smartglasses that #record and #transcribe #conversations, providing real-time information to the wearer. The glasses, priced at £199, aim to enhance intelligence and memory, but privacy concerns arise due to the lack of an external recording indicator. https://techcrunch.com/2025/08/20/harvard-dropouts-to-launch-always-on-ai-smart-glasses-that-listen-and-record-every-conversation/?eicker.news #tech #media #news

#Amazon is acquiring #Bee, a startup that makes a #wearable device using #AI to #transcribe #conversations and generate personalised summaries. Amazon will work with Bee to give users greater control over their devices and data. https://www.theverge.com/news/711621/amazon-bee-ai-wearable-acquisition?eicker.news #tech #media #news

#Amazon is acquiring #Bee, a startup that makes a #wearable device using #AI to #transcribe #conversations and generate personalised summaries. Amazon will work with Bee to give users greater control over their devices and data. https://www.theverge.com/news/711621/amazon-bee-ai-wearable-acquisition?eicker.news #tech #media #news

AWS Step Functionsで議事録自動化フローを作成しました（途中経過）
https://qiita.com/ManaTe4BOYS/items/cb10163149af3b59ac0f?utm_campaign=popular_items&utm_medium=feed&utm_source=popular_items

#qiita #AWS #初心者 #stepfunctions #Transcribe #bedrock

Evernote has gotten seriously better

Almost doubling the subscription price of a program is usually not a recipe for customer satisfaction, but I can now make an exception for Evernote. Yes, the note-taking app that many people seemed ready to leave for dead three and a half years ago, when a European software company I had never heard of bought the app that I’d already been relying on for more than a dozen years.

The Nov. 16, 2022 TechCrunch headline on Kyle Wiggers’ post about Bending Spoons buying Evernote practically sighed, calling it “the end of an era.” By then, Evernote had fallen greatly from its heights a decade earlier, when it was a frequent home-screen occupant of iPhones in Silicon Valley and visionary CEO Phil Libin talked about making the company a “100-year startup.”

After Evernote’s introduction in 2014 of a business-card-scanning feature that I still rely on, I had not seen many new features useful to my own work. And I was still experiencing too many note-synchronization glitches between my devices, despite the vow of new management to focus on the app’s core note-taking functions.

All this had me wondering if the premium account that had been a fixture in my Web-services budget since 2015, and which had increased from $45 a year to $69.99, was an expense I would be better off zeroing out. The obvious alternative was Microsoft’s OneNote, which I was already paying for with my Office 365 subscription.

But some genius at Microsoft elected to retire the company’s Evernote-to-OneNote importer in September of 2022, making any such migration a lot more difficult. And then the new management at Bending Spoons got to work improving the product–and one of the first things the people at that Milan-based firm addressed was note sync.

They made enough progress that when the company announced a steep rate hike at the end of 2023–from $69.99 to $129.99–I grudgingly decided to re-up for one more year and see where things stood. Eight months later, I was pleasantly surprised to see words I typed in my laptop’s copy of Evernote appearing my phone’s copy of the app a second later.

Today, that lag is barely discernible.

And last year, Evernote added a feature that directly helps one of my core tasks as a journalist, writing down what people say. This AI Transcribe tool that once served up long, unbroken blocks of text has gotten increasingly accurate and useful.

When I tested this Saturday afternoon by having the Evernote app on my iPad record and then transcribe a video of a 15-minute talk from the HumanX AI conference, the transcript that it generated in 40 seconds was just about as accurate as the Read AI transcript on the conference’s site (aside from botching a company name) and added bullet points and numbers to match the speaker’s pacing.

Evernote’s new management has also done a good job of communicating with its customers, posting detailed release notes for the app’s Windows, Mac, Android and iOS/iPadOS versions (why is that so hard for other companies?) to go with its frequent updates. The app still needs a word-count function, but overall it seems immensely improved from two years ago.

Meanwhile, Evernote’s cross-platform competition hasn’t done as much to earn my business.

OneNote still doesn’t have a built-in business-card-scanning feature–that requires a separate app–and sees fewer updates than Evernote, with less detail published about the content of these updates. I don’t see the same hustle at Microsoft.

And I just don’t want to trust this function to Google Keep, the free app Google had the temerity to announce in 2013 literally a week after the company killed off its beloved RSS client Google Reader. Twelve years later, a lot more of my digital life now happens on Google services, which makes me even less interested in handing over this extra bit of it to that company.

Yes, $129.99 is a serious amount of money for a Web app and service–but not in the context of one that I use and find useful multiple times a day.

#100YearStartup #AITranscribe #BendingSpoons #businessCards #Evernote #GoogleKeep #noteTaking #notes #OneNote #PhilLibin #transcribe #transcription

Evernote has gotten seriously better

Almost doubling the subscription price of a program is usually not a recipe for customer satisfaction, but I can now make an exception for Evernote. Yes, the note-taking app that many people seemed ready to leave for dead three and a half years ago, when a European software company I had never heard of bought the app that I’d already been relying on for more than a dozen years.

The Nov. 16, 2022 TechCrunch headline on Kyle Wiggers’ post about Bending Spoons buying Evernote practically sighed, calling it “the end of an era.” By then, Evernote had fallen greatly from its heights a decade earlier, when it was a frequent home-screen occupant of iPhones in Silicon Valley and visionary CEO Phil Libin talked about making the company a “100-year startup.”

After Evernote’s introduction in 2014 of a business-card-scanning feature that I still rely on, I had not seen many new features useful to my own work. And I was still experiencing too many note-synchronization glitches between my devices, despite the vow of new management to focus on the app’s core note-taking functions.

All this had me wondering if the premium account that had been a fixture in my Web-services budget since 2015, and which had increased from $45 a year to $69.99, was an expense I would be better off zeroing out. The obvious alternative was Microsoft’s OneNote, which I was already paying for with my Office 365 subscription.

But some genius at Microsoft elected to retire the company’s Evernote-to-OneNote importer in September of 2022, making any such migration a lot more difficult. And then the new management at Bending Spoons got to work improving the product–and one of the first things the people at that Milan-based firm addressed was note sync.

They made enough progress that when the company announced a steep rate hike at the end of 2023–from $69.99 to $129.99–I grudgingly decided to re-up for one more year and see where things stood. Eight months later, I was pleasantly surprised to see words I typed in my laptop’s copy of Evernote appearing my phone’s copy of the app a second later.

Today, that lag is barely discernible.

And last year, Evernote added a feature that directly helps one of my core tasks as a journalist, writing down what people say. This AI Transcribe tool that once served up long, unbroken blocks of text has gotten increasingly accurate and useful.

When I tested this Saturday afternoon by having the Evernote app on my iPad record and then transcribe a video of a 15-minute talk from the HumanX AI conference, the transcript that it generated in 40 seconds was just about as accurate as the Read AI transcript on the conference’s site (aside from botching a company name) and added bullet points and numbers to match the speaker’s pacing.

Evernote’s new management has also done a good job of communicating with its customers, posting detailed release notes for the app’s Windows, Mac, Android and iOS/iPadOS versions (why is that so hard for other companies?) to go with its frequent updates. The app still needs a word-count function, but overall it seems immensely improved from two years ago.

Meanwhile, Evernote’s cross-platform competition hasn’t done as much to earn my business.

OneNote still doesn’t have a built-in business-card-scanning feature–that requires a separate app–and sees fewer updates than Evernote, with less detail published about the content of these updates. I don’t see the same hustle at Microsoft.

And I just don’t want to trust this function to Google Keep, the free app Google had the temerity to announce in 2013 literally a week after the company killed off its beloved RSS client Google Reader. Twelve years later, a lot more of my digital life now happens on Google services, which makes me even less interested in handing over this extra bit of it to that company.

Yes, $129.99 is a serious amount of money for a Web app and service–but not in the context of one that I use and find useful multiple times a day.

#100YearStartup #AITranscribe #BendingSpoons #businessCards #Evernote #GoogleKeep #noteTaking #notes #OneNote #PhilLibin #transcribe #transcription

Since I am one of the three people left who can read cursive, I volunteered to help the National Archives transcribe hand-written pension files from the Revolutionary War. It's a pretty cool diversion during the day and it only takes a few minutes.

https://www.archives.gov/citizen-archivist/missions/revolutionary-war-pension-files

#cursive #transcribe #transcription #pension #RevolutionaryWar #history #nationalarchives

Since I am one of the three people left who can read cursive, I volunteered to help the National Archives transcribe hand-written pension files from the Revolutionary War. It's a pretty cool diversion during the day and it only takes a few minutes.

https://www.archives.gov/citizen-archivist/missions/revolutionary-war-pension-files

#cursive #transcribe #transcription #pension #RevolutionaryWar #history #nationalarchives

WhatsApp Voice Message Transcripts converts voice into text #apps #digitallife #featured #meta #news #socialmedia #transcribe #transcript #whatsapp #whatsappvoicetotext #whatsappvoicetranscripts

https://soyacincau.com/2024/11/22/whatsapp-voice-message-transcripts-how-to-activate/

Kiwi "company has been developing artificial intelligence (#AI) tools that #transcribe te #reo Māori" - but: Why Google Maps can't pronounce Māori place names https://www.rnz.co.nz/news/te-manu-korihi/531303/why-google-maps-can-t-pronounce-maori-place-names

Kiwi "company has been developing artificial intelligence (#AI) tools that #transcribe te #reo Māori" - but: Why Google Maps can't pronounce Māori place names https://www.rnz.co.nz/news/te-manu-korihi/531303/why-google-maps-can-t-pronounce-maori-place-names

This one is for the #musicians, Andy Robinson's #Transcribe is an audio workstation specifically for transcribing music recordings, built for Mac, Linux, and that other OS (fenestra? porthole? jalousie?).

I bought my lifetime subscription a very long time ago, several computers ago, and I'm still amazed every time I get Andy's emails and sure enough, my license is still good after all these years.

Here are some extracts from Transcribe! reviews and emails from users:

https://www.seventhstring.com/xscribe/reviews.html

This one is for the #musicians, Andy Robinson's #Transcribe is an audio workstation specifically for transcribing music recordings, built for Mac, Linux, and that other OS (fenestra? porthole? jalousie?).

I bought my lifetime subscription a very long time ago, several computers ago, and I'm still amazed every time I get Andy's emails and sure enough, my license is still good after all these years.

Here are some extracts from Transcribe! reviews and emails from users:

https://www.seventhstring.com/xscribe/reviews.html

New post: it takes ~3 mins to transcribe an hour of podcast episode. Here's how you can do it with Whisper.cpp and Poddl: https://codito.in/transcribe-podcasts-with-whisper/

#ai #podcast #whisper #poddl #speechtotext #script #transcribe #automation

263 Creative Marketing Unleashed and iOS 17's Impact on Podcasting - https://pca.st/vovwb13q Tip how @elsie uses #Matter to #transcribe #Podcasts

Practicing #transcription today, I'm using an audio of a #speech given by Hilliary Clinton at a world conference for women. Beautiful speeches like this one use more complex sentences, delivered slowly and clearly. I'm sure Hillary Clinton didn't write all her speeches, but she is a brain trust and an amazing speaker. And what a beautiful thing to take down these words. Imagine trying to #transcribe the speeches of the MAGA OG.

Practicing #transcription today, I'm using an audio of a #speech given by Hilliary Clinton at a world conference for women. Beautiful speeches like this one use more complex sentences, delivered slowly and clearly. I'm sure Hillary Clinton didn't write all her speeches, but she is a brain trust and an amazing speaker. And what a beautiful thing to take down these words. Imagine trying to #transcribe the speeches of the MAGA OG.

"In addition to text, #ToxicityDetection uses speech cues such as tones and pitch to hone in on toxic intent in speech."

Good luck AWS, my toxicity is completely deadpan

Flag harmful language in spoken conversations with #Amazon #Transcribe #Toxicity #Detection | #AWS #MachineLearning Blog https://aws.amazon.com/blogs/machine-learning/flag-harmful-language-in-spoken-conversations-with-amazon-transcribe-toxicity-detection/

#speechRecognition #AI #moderation

"In addition to text, #ToxicityDetection uses speech cues such as tones and pitch to hone in on toxic intent in speech."

Good luck AWS, my toxicity is completely deadpan

Flag harmful language in spoken conversations with #Amazon #Transcribe #Toxicity #Detection | #AWS #MachineLearning Blog https://aws.amazon.com/blogs/machine-learning/flag-harmful-language-in-spoken-conversations-with-amazon-transcribe-toxicity-detection/

#speechRecognition #AI #moderation

Some #FavoriteTools

#perplexity.ai for thoughtful web queries. (Shows reasoning and references.)

#Vim for #TextEditing

#Firefox w/ #Tridactyl #uBlockOrigin

#Raindrop for #Tagging and #Bookmarking web pages

#Inoreader #RSS reader

#Pandoc for document creation/conversion

#Ubuntu #Linux (even if it's running under Windows #wsl

#Python w/ #Jupyter #Pandas #Matplotlib #numpy #scipy #plac

#TheBrain for #KnowledgeManagement (but considering moving to #Obsidian)

#MoonReader for ebooks and PDFs

#RolandFP90 and A-88 for #piano

#transcribe for working through #music #video

#Leatherman #SkeletoolCX for EDC multitool

#Benchmade #Griptilian when I need more of an actual knife

#Signal for #messaging

#mastodon for social media

#antennapod for podcasts

#streamlightwedgext for #EDC flashlight

Some #FavoriteTools

#perplexity.ai for thoughtful web queries. (Shows reasoning and references.)

#Vim for #TextEditing

#Firefox w/ #Tridactyl #uBlockOrigin

#Raindrop for #Tagging and #Bookmarking web pages

#Inoreader #RSS reader

#Pandoc for document creation/conversion

#Ubuntu #Linux (even if it's running under Windows #wsl

#Python w/ #Jupyter #Pandas #Matplotlib #numpy #scipy #plac

#TheBrain for #KnowledgeManagement (but considering moving to #Obsidian)

#MoonReader for ebooks and PDFs

#RolandFP90 and A-88 for #piano

#transcribe for working through #music #video

#Leatherman #SkeletoolCX for EDC multitool

#Benchmade #Griptilian when I need more of an actual knife

#Signal for #messaging

#mastodon for social media

#antennapod for podcasts

#streamlightwedgext for #EDC flashlight

PRODUCTHEAD: How to transcribe and analyse your user research recordings

This week’s #productmanagement newsletter by Jock Busuttil (@jock) features @maria_rosala @mulegirl @KaraAnn @uxdesigninst

#prodmgmt #userinterviews #transcribe #analyze https://imanageproducts.com/producthead-how-to-transcribe-and-analyse-your-user-research-recordings/

PRODUCTHEAD: How to transcribe and analyse your user research recordings

This week’s #productmanagement newsletter by Jock Busuttil (@jock) features @maria_rosala @mulegirl @KaraAnn @uxdesigninst

#prodmgmt #userinterviews #transcribe #analyze https://imanageproducts.com/producthead-how-to-transcribe-and-analyse-your-user-research-recordings/

I'm never gonna do transcriptions again
Guilty fingers have got no rhythm
Though it's easy to pretend
I know you're not a fool (computer)

- @Meyerweb / George Michael duet. #transcribe #software
https://meyerweb.com/eric/thoughts/2023/03/23/peerless-whisper/

I'm never gonna do transcriptions again
Guilty fingers have got no rhythm
Though it's easy to pretend
I know you're not a fool (computer)

- @Meyerweb / George Michael duet. #transcribe #software
https://meyerweb.com/eric/thoughts/2023/03/23/peerless-whisper/

I found out today that #OpenAI was funded by a guy I don't like to support. It's a shame because I was liking some of the tools for transcription and translation that are local and not "cloud". Any similar #AI tools in development that you know that can be used locally to #translate, #transcribe? #Finnish would be great

Do any devs out there know if you can tell the #OpenAI #whisper #transcribe system to use British English spelling in transcriptions? I see "--language en" is an option (and the one I use). There is a "uk", but that's Ukrainian...

The reason I ask is to prevent misspellings of common terms like "color/colour", "favorite/favourite", etc...