Any YouTube video or podcast episode: the full transcript with timestamps, notes, chapters, and answers that come from the transcript itself. Free to start: 30 AI credits a month.
Mario Nawfal and Chris Martenson discuss a wave of Houthi strikes on Saudi oil, gas and airport facilities, including a two-day fire at the Khurais processing plant tracked via NASA FIRMS satellite data, and a ship strike in the Strait of Hormuz filmed by a drone. They also cover a shaky Trump-brokered deal with Putin to halt strikes on energy infrastructure, tanker shortages, diesel pressure on Trump, and Chris's argument that gold, silver and oil markets are manipulated to suppress price signals. Chris concludes there is no credible military or diplomatic solution on Iran and that the world is heading into an energy and food emergency.
Chapters
00:00Saudi facilities struck and Strait of Hormuz mine
Dear Therapists with Lori Gottlieb and Guy Winch · 49 min
Summary
In this encore episode, therapists Lori Gottlieb and Guy Winch counsel Shreena, who is grieving a nine-month breakup with Joe after ending the relationship herself. They explore how her unresolved childhood pain and high tolerance for unacceptable behavior led her to mold herself for partners and treat crumbs of affection as full meals. Shreena later reports that she began weekly therapy, chose distance from Joe over reuniting, and is learning to sit with difficult emotions.
In this conversation on Point of Relation, storyteller and mythologist Michael Meade tells Thomas Hübl about being drafted for the Vietnam War at 19, refusing to fight, and enduring months of solitary confinement and a hunger strike that nearly killed him. He describes how mythological figures visited him in his cell and how the experience became an initiation that shaped his later work with at-risk youth, prisoners, and refugees. Meade argues that humanity is collectively in a liminal rite of passage, having lost the old world but not yet found the new one, and that change happens through individual souls, not at the collective level. He closes with the insight that permission and forgiveness for who and where we are form the starting place for finding the path with heart.
In dieser Folge des ARD-Podcasts „Mal angenommen“ durchspielen Marcel Heberlein und Birte Sennigsen das Gedankenexperiment, was passieren würde, wenn der Staat in Deutschland flächendeckend Überwachungskameras mit Gesichtserkennung einsetzt. Grundlage sind das Pilotprojekt am Berliner Bahnhof Südkreuz, Gespräche mit einem Gewerkschafter der Polizei, einer Ethikerin und einem Sicherheitsforscher. Sie kommen zum Schluss: Im besten Fall hilft die Technik bei Fahndung und Sicherheitsgefühl, im schlechtesten Fall entsteht ein Überwachungsstaat mit fehleranfälliger KI und missbräuchlicher Datennutzung.
Marc Andreessen interviews Replit founder Amjad Masad about AI-driven software creation, how Replit's agents build and deploy apps from plain English prompts, and why coding is advancing faster than other AI domains thanks to verifiable reinforcement learning. They debate whether we are on track to AGI, with Masad arguing that 'good enough' current AI creates a local maximum trap that relieves pressure to solve general intelligence. Masad also tells his life story, from a childhood in Amman, Jordan to hacking his university grades and eventually founding Replit.
30 Minutes to President's Club | No-Nonsense Sales · 21 min
Summary
This episode walks through a cold calling script that can book one meeting for every three connects. The host breaks the call into three parts: the 'heard the name tossed around' opener, a problem-based pitch, and the 'Mr. Miyagi' method for handling objections. The core theme is doing the opposite of typical telemarketers at every step.
Chapters
00:00Intro and Sam McKenna's social selling course
Elise Loehnen talks with mythologist Michael Meade about the crisis of masculinity, the manosphere, and what he sees as the root cultural diagnosis: a lack of soul. Meade argues that the hero's journey has been misused as the only story, that Trump is a destructive, unconscious trickster rather than a healer, and that the world changes when individuals grow their own souls. He also describes how the Catholic Church removed the soul as a third force at a council in Constantinople in 869, which he sees as the origin of modern polarization. The path forward, he says, is each person awakening their innate genius and following what they love.
Chapters
00:00Introduction and the crisis of masculinity
02:50Lack of soul as the diagnosis
05:52The hero's journey fallacy and the genius myth
11:43Initiation, wounds, and the manosphere as adolescence
این قسمت اول از «رادیو کار نکن» گفتگویی است با سلمان قاسمی، بنیان‌گذار زاگرس کمپرسور، درباره مسیر شغلی و تجربه‌های کاریش از کودکی در ایلام تا راه‌اندازی شرکت خودش. او از کار در آریا تکنیک نزد مهندس فیض‌آبادی و سپس کار در حوزه کمپرسورسازی می‌گوید و توضیح می‌دهد چگونه دو سال و نیم پیش با همسرش کسب‌وکار مستقل خود را با قرارداد اول ۴۹ میلیونی شروع کرد و امروز به تیمی حدود پانزده نفر رسیده است. پیام اصلی او اهمیت علاقه به کار، احساس مسئولیت، نگاه بلندمدت به جای منافع کوتاه‌مدت و مشاوره صادقانه به مشتری است. بخش پایانی ضبط به دلیل افت کیفیت متن، نامفهوم و پراکنده است.
Sir Ken Robinson argues that creativity is as important as literacy in education and should be treated with the same status. He contends that children are naturally creative and unafraid of being…
A YouTube video, or one episode from Spotify, Apple Podcasts or a podcast feed. A direct link to an audio or video file works too.
2
The words are fetched or transcribed
A video with captions takes seconds. Without captions the audio is transcribed, which takes about a minute for every 15 to 20 minutes.
3
Read it, download it, share it
The summary, chapters and notes open on their own page. A link someone has already transcribed opens at once.
Before you paste
Up to 90 minutes of audio per link. YouTube videos with captions can run longer.
One video or episode per link for free. Whole YouTube playlists are a paid workspace feature; channels and live streams are not supported.
Private, members-only and age-restricted videos cannot be fetched, and shows that are only on Spotify have no public feed to transcribe.
Free transcripts are timestamped paragraphs. Speaker labels (who said what) are a paid workspace feature.
Transcript pages are public. Do not paste links to anything private.
The summary and notes are written by AI from the transcript. Check anything important against the recording.
Questions
How do I get the transcript of a YouTube video?
Copy the video's link, paste it into the box at the top of this page and press Transcribe. If the video has captions the transcript is ready in a few seconds; if not, the audio is transcribed, which takes about a minute for every 15 to 20 minutes of video.
Which links work?
YouTube videos, Spotify and Apple Podcasts episodes, podcast feed items, and direct links to audio or video files. Spotify does not share audio, so the same episode is found in the show's public podcast feed; shows that are only on Spotify have no such feed and cannot be transcribed.
Is it really free?
Free to start, with no card, and the file tools are free. A free account comes with 30 AI credits, and 30 more every month: enough for about 10 videos with captions, or 2.5 hours of Apple Podcasts or Spotify audio, as public transcripts. Reading transcript pages is free for everyone. For more, Blumify Unlimited (US$9.99 a month, or US$59.99 a year) covers transcripts, notes, questions and translations without counting credits: videos with captions without a limit, and 50 hours of audio a month at full speed, then up to 1 hour a day.
How long can a video or episode be?
Up to 90 minutes when the audio has to be transcribed for free, and up to 4 hours for a private transcript in your workspace (it uses AI credits). YouTube videos with captions can be longer. Uploaded files can be up to 10 minutes and 100MB.
Who can see my transcript?
Free transcripts from links are public: each gets a page with its summary and notes, so a link someone else already transcribed opens instantly. To keep one to yourself, make it in your private workspace instead (it uses AI credits). Files you upload are never published, and they are deleted after processing.
How accurate is it?
We measured it on podcasts, talks and earnings calls with human transcripts: transcripts on our standard speech model were 87.5% accurate, and the most accurate model we tested reached 90.1% (with bought credits or Unlimited, you can choose it for a private transcript). Videos with captions use them, and YouTube's automatic captions were 94.3% accurate. Full results: blumify.io/accuracy.
What languages does it handle?
Most common languages are detected automatically, including English, Spanish, Portuguese, German, French, Italian, Dutch, Polish, Russian, Japanese, Korean, Chinese, Hindi and Arabic. Quality varies by language and by recording.
Can it tell speakers apart?
Yes, on private transcripts: speaker labels (who said what) are a paid feature of the workspace. Public transcripts come as timestamped paragraphs without names.
What happens to the files I upload?
They go to private storage that is never public and never listed. Only your own session can read, cancel or download the job. The original is deleted as soon as processing succeeds, and the result stops being served about 90 minutes later.