Whisper, every mention
43 scenes, the whole family · ← back to Whisper
tap a year for its mentions
every year anyone Vibhu Sapra 26Alessio Fanelli 10Shawn Wang 9Andrew Hsu 8Pavan Kumar Reddy 2Simon Willison 1George Hotz 1Ethan Sutin 1Deedy Das 1
Verbatim, from the transcripts: the passages where Whisper comes up
The Future of AI Infra: from Kubernetes to Agent Sandboxes — Akshat Bubna, Modal CTO
- ▶ 49:32 unnamed speaker There's Qtai TTS, customizing, oh, Chatterbox, um, you know, there was a customizing whisperer.
Mistral: Voxtral TTS, Forge, Leanstral, & Mistral 4 — w/ Pavan Kumar Reddy & Guillaume Lample
- ▶ 26:34 Pavan Kumar Reddy And I think, uh, a big people, I think there's a big, uh, uh, rich ecosystem of, uh, people finding whisper and people want the same thing with Voxer. 2 times in the scene
- ▶ 29:15 unnamed speaker But a big thing, so Whisper is known for 32nd generation, uh, 32nd processing. 3 times in the scene
Anthropic, Glean & OpenRouter: How AI Moats Are Built with Deedy Das of Menlo Ventures
Personalized AI Language Education — with Andrew Hsu, Speak
- ▶ 5:06 Andrew Hsu Before 20, 22, when Whisper came out, when ChatGPT came out in the years, 2 times in the scene
- ▶ 13:18 Andrew Hsu I guess, like, phase two was really, twenty-twenty-two when LLMs came out, and Whisper came out, and that allowed us to go from this more supplemental speaking practice tool to more full-featured language tutoring where we could use LLMs,…
- ▶ 23:40 Andrew Hsu So that was also when Whisper dropped. 5 times in the scene
- ▶ 26:16 Alessio Fanelli Or did you almost feel like, okay, we spent all this money and time building these models and now we're just going to use whisper. 2 times in the scene
Snipd: The AI Podcast App for Learning — with CEO Kevin Ben-Smith
- ▶ 6:07 Shawn Wang Before Whisper. 2 times in the scene
- ▶ 31:13 Shawn Wang What was the sort of before Whisper? 2 times in the scene
- ▶ 56:28 Shawn Wang You know, you have whisper, you have a pipeline and everything. 2 times in the scene
DeepSeek V3, SGLang, and the state of Open Model Inference in 2025 (Quantization, MoEs, Pricing)
- ▶ 47:44 unnamed speaker Again, this is generally done by our customers before they come to us for inference, but there's examples like in the healthcare world, fine-tuning models for understanding medical jargon, like for Whisper, for instance, you know, a… 2 times in the scene
2024 Year in Review: The Big Scaling Debate, the Four Wars of AI, Top Themes and the Rise of Agents
- ▶ 1:10:12 Shawn Wang The batch transcription, I would say whisper is, is, uh, something that you should be using on a, on a, as much as possible and then code generation and kind of solve, uh, there's, there's different tiers of code generation
[Paper Club] Molmo + Pixmo + Whisper 3 Turbo - with Vibhu Sapra, Nathan Lambert, Amgadoz
- ▶ 17:57 Vibhu Sapra Was that like, you know, whisper or something or? 3 times in the scene
- ▶ 48:33 Vibhu Sapra Um, and then we've got a little update on Whisper. 4 times in the scene
- ▶ 50:22 Vibhu Sapra Whisper. 13 times in the scene
- ▶ 52:01 unnamed speaker Uh, for example, in the Whisper Large VIII, it has 32 layers in the encoder and 32 layers in the decoder. 5 times in the scene
- ▶ 52:20 unnamed speaker And then later on, they added large V two and large V three.
- ▶ 57:58 Vibhu Sapra The difference with Whisper Turbo and Distal Whisper is primarily the pruning versus distillation. 6 times in the scene
- ▶ 59:04 unnamed speaker It uses exactly the same data set for whisper large V three, except for the translation section, which is, uh, excluded.
- ▶ 1:00:05 unnamed speaker And I believe, like, if you train a single model on multiple datasets, multiple domains, multiple languages, it becomes more robust, and this is like the premise behind Whisper. 2 times in the scene
- ▶ 1:04:16 unnamed speaker Uh, so the question is like, is the, the, the whisper model fast enough to be real time? 4 times in the scene
- ▶ 1:04:36 unnamed speaker If you've got the parameters correctly and if you size it correctly, and this is even applicable with like the large V two large V three. 2 times in the scene
- ▶ 1:04:36 unnamed speaker If you've got the parameters correctly and if you size it correctly, and this is even applicable with like the large V two large V three. 2 times in the scene
- ▶ 1:08:37 unnamed speaker So, like, whisper spawn,
- ▶ 1:08:39 unnamed speaker Base.
- ▶ 1:08:39 unnamed speaker Tiny.
- ▶ 1:08:40 unnamed speaker Medium.
- ▶ 1:08:41 unnamed speaker Large.
- ▶ 1:10:35 unnamed speaker Uh, but the other models are, like, kind of similar, except maybe large v three can be a hit or miss. 3 times in the scene
- ▶ 1:12:45 unnamed speaker Uh, always excited by your whisper updates and explanations.
Building AGI in Real Time (OpenAI Dev Day 2024)
- ▶ 1:01:03 Simon Willison Because, yeah, often I want to do, I do a lot of work with, like, transcripts of hour-long YouTube videos, which currently I run them through Whisper
Building AGI with OpenAI's Structured Outputs API
- ▶ 58:12 Shawn Wang Batch, Vision, Whisper, and then Team Enterprise stuff.
- ▶ 1:01:52 Alessio Fanelli Let's talk about my favorite model, Whisper. 8 times in the scene
The Winds of AI Winter (Q2 Four Wars of the AI Stack Recap)
- ▶ 46:59 unnamed speaker I was kind of bearish on conformers because I looked at the state of existing conformer research in ICM, Eclair, and, and NeurIPS, and they were far, far, far behind Whisper, mostly because of scale, like the, the sheer amount of resources… 2 times in the scene
Breaking down the OG GPT Paper by Alec Radford
- ▶ 0:25 unnamed speaker I've done some posts about whisper.
- ▶ 31:23 unnamed speaker Uh, so they are trying to create a multitask format, and this is similar to like what people have used in feature work like T five and whisper, where basically you're trying to, uh, model different tasks, just using tokens and special…
Personal AI Meetup - Bee, BasedHardware, LangChain LangFriend, Deepgram EmilyAI
- ▶ 27:11 Ethan Sutin It was very complicated, um, like, because we were using, like, local whisper, local models, and, like, getting it to work on CUDA, Mac, Windows.
- ▶ 38:07 unnamed speaker Of, uh, uh, speech from the opening, I, uh, whispered endpoint, and it says, if it's checking, uh, yeah, there we go.
A Brief History of the Open Source AI Hacker - with Ben Firshman of Replicate
Building an open AI company - with Ce and Vipul of Together AI
- ▶ 22:40 Shawn Wang I've, um, I think a lot of people think that, um, there's, there's a theory that Whisper was released so that you could transcribe YouTube and then use that as a source of tokens.
Ep 18: Petaflops to the People — with George Hotz of tinycorp
- ▶ 15:51 George Hotz I've a rewrite of whisper.