#whisper
#TIL that some #SpeechToText models interpret audio noise as [applause]. 
So you know you've messed things up, if you speak to your self-developed Speech-To-Text wrapper and it outputs: [applause]. 
Like "Hey you've messed things up!" *Clap, clap, clap*
Not sure I like this kind of encouragement...
Whispers From The Void, Transcribed With AI - ‘Hearing voices’ doesn’t have to be worrisome, for instance when software-defined ... - https://hackaday.com/2025/08/08/whispers-from-the-void-transcribed-with-ai/ #artificialintelligence #digitalaudiohacks #radiohacks #whisper #openai #audio #radio #cuda #gpu #sdr #vad
Hacker News Show HN: I built a tool to replace capcut audio transcription
https://meetcosmos.com/free-audio-transcription/
#ycombinator #audio_transcription #speech_to_text #AI_transcription #free_transcription #whisper #browser_transcription
Using Home Assistant OS, I wired up Whisper for speech recognition, Piper for voice responses, and Ollama for the LLM brain — all running on my own machines, stitched together with the Wyoming protocol. I could literally talk to my smart home, and it talked back. All offline, all private.
Sometime back, I got obsessed with the idea of running my own voice assistant — fully local, no Google, no Amazon, no "cloud intelligence." Just me, my wife and our server(s)...
So I built Marvin.
Bug report über Whisper, das LLM-basierte Transkriptionstool:
Anscheinend wird Stille am Ende auf Deutsch transkribiert als
"Untertitelung des ZDF für funk, 2017."
https://github.com/openai/whisper/discussions/2608#discussioncomment-13790984
säggsisch Transkript nacharbeiten macht Garnelen spaß
...eben weil du regelmäßig so Dinge wie "Garnelen" zu "gar keinen" Spaß korrigieren musst #whisper #soziologie #qualitativeresearch
How To Train A New Voice For Piper With Only A Single Phrase - [Cal Bryant] hacked together a home automation system years ago, which more recent... - https://hackaday.com/2025/07/09/how-to-train-a-new-voice-for-piper-with-only-a-single-phrase/ #artificialintelligence #speechsynthesis #texttospeech #chatterbox #pipervoice #pytorch #whisper #gpu
Si vous êtes chercheur.se dans le domaine des SHS, vous pouvez utiliser #Whisper pour transcrire des enregistrements dans #ShareDocs, durée de traitement selon le fichier soumis et le nombre de dépôts simultanés : 1 à 24 heures. https://documentation.huma-num.fr/sharedocs-traitement/#speechtotext-whisper
NVIDIAの自動音声認識モデルParakeetをサポートし、最新のMacで最大300倍高速に英語の文字起こしが可能になったAI文字起こしアプリ「MacWhisper v12.12」がリリース。
https://applech2.com/archives/20250630-macwhisper-support-nvidia-parakeet.html
Franchement c'est bEnger !!! 
(Ça fait de la transcription audio)
https://f-droid.org/packages/com.module.notelycompose.android
Whisper has a serious challenger: Moshi STT
Developed by the French research lab Kyutai, Moshi STT is a new open-source speech recognition system that’s blazingly fast, highly accurate, and optimized for Apple Silicon and CUDA — all designed with real-time performance in mind.
... erster Eindruck und Spoiler: wertvolle Impulse und Diskurs gingen weit über grundlegende KI-Tool-Theorie und allgemeine Nutzungserfahrungen hinaus - wie etwa die automatische Audiotranskription im @TIB_AVPortal oder den praktischen Erfahrungen des Filminstituts Hannover mit mit #Whisper. Aspekte reichten von praktischer KI-Softwarenutzung zur #Videoproduktion bis hin zu kritischer Auseindersetzung mit Prozessen, Rechtsfragen und Herausforderungen für Filmbibliotheken- und Archive ...
#Whisper #WebGPU by #Huggingface sounds very exciting!
Does this mean an #activitypub server could delegate translation-into-user's-language of all the posts to the user's device?
I'm too thick to have been able to find any system-requirements information for just the text-translation feature... Is this #translation feature likely to fly on mobile devices too?
Am I getting too excited too soon?
https://dev.to/proflead/real-time-audio-to-text-in-your-browser-whisper-webgpu-tutorial-j6d
https://github.com/keatonkraiger/Whisper-Transcribe-and-Translate-Tutorial
#Apple’s new #SpeechAPIs, #SpeechAnalyzer and #SpeechTranscriber, are significantly faster than #OpenAI’s #Whisper for #transcription. https://www.macstories.net/stories/hands-on-how-apples-new-speech-apis-outpace-whisper-for-lightning-fast-transcription/?eicker.news #tech #media #news
Just to clarify: I don't think #AI use is inherently bad for science.
#LLM‘s can help you reword, make text flow better, be more precise and write better, because – unfortunately – training data also includes lots of good scientific texts.
ASR systems like #whisper allow you to spend less time on word by word #transcription and more on what's between the lines.
But use for citing literature? Writing whole sections or papers? Review? Coding in qualitative research?!
That's an issue
люди предлагали выкладывать субтитры Whisper’а в публичный доступ, что я и сделала!
https://wonderfox.anyaforger.art/subtitles/en/
на данный момент есть субтитры к What’s with Andy (кроме первого сезона и части третьего - случайно утеряла часть субтитров при удалении папки с мультиком). этот пост будет обновляться с появлением новых субтитров (а они будут)