Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Introducing Canto, our latest speech model for real-time dictation. On our evaluation of 9.8 hours of English Flow dictations from more than 2,300 speakers, Canto achieved a 3.4% word error rate—the lowest among the models we tested. Learn more about Canto from our CSO, Ariya Rastrow:

WisprFlow's profile picture

Wispr Flow

34,323 subscribers

16,464 Aufrufe • vor 9 Tagen •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

16 Kommentare

Profilbild von Wispr Flow
Wispr Flowvor 9 Tagen

Learn more about Canto here:

Profilbild von Artem Russakovskii
Artem Russakovskiivor 9 Tagen

Is it already live in Wispr now?

Profilbild von Edward C
Edward Cvor 9 Tagen

y'all can't call bro a wrapper no more

Profilbild von 𝕏 frjps 🇻🇦⚡️ 🐟
𝕏 frjps 🇻🇦⚡️ 🐟vor 8 Tagen

The experience is terrible on the iPhone and absent on the iPad.

Profilbild von Jack Sondag
Jack Sondagvor 8 Tagen

Will this be the new default model used or will users decide?

Profilbild von Harshit Khemani
Harshit Khemanivor 9 Tagen

@Scobleizer chief speech officer 😭⁉️

Profilbild von Yaya Soumah
Yaya Soumahvor 9 Tagen

@Scobleizer So is live or not?

Profilbild von Mac
Macvor 8 Tagen

@Scobleizer @grok how do i select Canto model in wispr flow macos app?

Profilbild von DeDi
DeDivor 8 Tagen

实时听写语音模型把错误率压到3.4%，听写真能替代打字

Profilbild von Shreekumar
Shreekumarvor 9 Tagen

no api for canto ?

Profilbild von Bonny Scotland
Bonny Scotlandvor 8 Tagen

Until Canto can pick up and transcribe Scottish correctly it will still be a terrible model.

Profilbild von Alek
Alekvor 9 Tagen

3.4% WER on 2300 speakers — yep.

Profilbild von Jorge Porras
Jorge Porrasvor 9 Tagen

Esta herramienta es la que más ha estado en mi top 5 de recomendadas por 2 años consecutivos. Háganse un favor y usen wisprflow.

Profilbild von Rimas
Rimasvor 9 Tagen

which models did you test canto against?

Profilbild von Rachit Kumar
Rachit Kumarvor 9 Tagen

those em-dashes @tankots 🙃

Profilbild von Patrick Gantner
Patrick Gantnervor 8 Tagen

Is Canto already avaliable for german language and rolled out to the users??

Ähnliche Videos

It's time we borrow an idea from our British friends. My proposal for a shadow cabinet isn’t just about holding our political leaders accountable - it strengthens and protects our democracy. Learn more from my speech today on the House floor.⬇️
1:45

It's time we borrow an idea from our British friends. My proposal for a shadow cabinet isn’t just about holding our political leaders accountable - it strengthens and protects our democracy. Learn more from my speech today on the House floor.⬇️

Congressman Wiley Nickel

154,690 Aufrufe • vor 1 Jahr

Today we're announcing $280M in Series B funding at a $2B valuation, led by our long-time investor and partner, Menlo Ventures. When we announced our Series A last May, most conversations I had about voice began with someone explaining why they doubted it. I don't have those conversations anymore. People tell me instead how much time they save talking instead of typing and what they want us to build next. In a little over a year, voice has gone from something people were questioning to something they rely on, and it happened faster than we expected. That shift is why we've raised a new round of funding. This funding represents a deeper investment in our products, lab, models, and our team. Alongside the funding, we're announcing a preview of our first proprietary speech model, Canto. Canto is a 2B parameter speech model trained for the places people actually talk: loud rooms, windy streets, a toddler in the background, a second language mixed into the first. Models like this will change how we all interact with devices, and Canto is the first in a rapid line of them. Each one will be larger and more capable than the last, and each should improve Flow's dictation in a way you can feel the day it ships. Larger speech models also open the door to the vision we dreamed up 5 years ago: using your voice as the primary way you interact with devices. That still has to be invented. A voice interface people rely on all day, in every app, doesn't exist anywhere yet. It's time for that to change. To Sahaj Garg, our team, and every person who gave Wispr Flow a chance - this one's for you.
1:00

Today we're announcing $280M in Series B funding at a $2B valuation, led by our long-time investor and partner, Menlo Ventures. When we announced our Series A last May, most conversations I had about voice began with someone explaining why they doubted it. I don't have those conversations anymore. People tell me instead how much time they save talking instead of typing and what they want us to build next. In a little over a year, voice has gone from something people were questioning to something they rely on, and it happened faster than we expected. That shift is why we've raised a new round of funding. This funding represents a deeper investment in our products, lab, models, and our team. Alongside the funding, we're announcing a preview of our first proprietary speech model, Canto. Canto is a 2B parameter speech model trained for the places people actually talk: loud rooms, windy streets, a toddler in the background, a second language mixed into the first. Models like this will change how we all interact with devices, and Canto is the first in a rapid line of them. Each one will be larger and more capable than the last, and each should improve Flow's dictation in a way you can feel the day it ships. Larger speech models also open the door to the vision we dreamed up 5 years ago: using your voice as the primary way you interact with devices. That still has to be invented. A voice interface people rely on all day, in every app, doesn't exist anywhere yet. It's time for that to change. To Sahaj Garg, our team, and every person who gave Wispr Flow a chance - this one's for you.

Tanay Kothari

615,867 Aufrufe • vor 1 Monat

Introducing a small portion of proprietary phenomics data from Recursion into the training of MolGPS, our foundation model for molecular property prediction, led to massive performance increases on new & existing benchmarks. Learn more about the impact of data & scale.
1:46

Introducing a small portion of proprietary phenomics data from Recursion into the training of MolGPS, our foundation model for molecular property prediction, led to massive performance increases on new & existing benchmarks. Learn more about the impact of data & scale.

Valence Labs

10,100 Aufrufe • vor 2 Jahren

Introducing Universal-1: Our most powerful and accurate Speech AI model yet. ✅ Trained on 12.5M hours of multilingual speech data ✅ 13.5% more accurate than models like Whisper ✅ Up to 30% fewer hallucinations than seq2seq models ✅ Just 38 seconds to process 1 hour of audio
1:01

Introducing Universal-1: Our most powerful and accurate Speech AI model yet. ✅ Trained on 12.5M hours of multilingual speech data ✅ 13.5% more accurate than models like Whisper ✅ Up to 30% fewer hallucinations than seq2seq models ✅ Just 38 seconds to process 1 hour of audio

AssemblyAI

23,511,312 Aufrufe • vor 2 Jahren

Introducing our latest campaign featuring our brand partners - starting with Manchester City 🎥 The ‘my fabric' video series reminds us that greatness cannot be achieved without belief, resilience, and conviction. You’ll learn about our brand ambassadors and their stories of grit to fortune. Our CMO haider shares more about the series: We begin with star midfielder Jack Grealish’s story ⤵
1:25

Introducing our latest campaign featuring our brand partners - starting with Manchester City 🎥 The ‘my fabric' video series reminds us that greatness cannot be achieved without belief, resilience, and conviction. You’ll learn about our brand ambassadors and their stories of grit to fortune. Our CMO haider shares more about the series: We begin with star midfielder Jack Grealish’s story ⤵

OKX

5,243,783 Aufrufe • vor 2 Jahren

Yesterday we introduced SeamlessExpressive — a new model that preserves unique vocal styles & expression for speech translation, built on our SeamlessM4T v2 foundation model. More details on the family of Seamless Communication models ➡️
2:36

Yesterday we introduced SeamlessExpressive — a new model that preserves unique vocal styles & expression for speech translation, built on our SeamlessM4T v2 foundation model. More details on the family of Seamless Communication models ➡️

AI at Meta

189,972 Aufrufe • vor 2 Jahren

As we celebrate our 100th birthday, MSHSAA is under attack from our state legislature. It's time to fight for the future of high school sports and activities in our state. Learn More:
0:32

As we celebrate our 100th birthday, MSHSAA is under attack from our state legislature. It's time to fight for the future of high school sports and activities in our state. Learn More:

MSHSAA

100,759 Aufrufe • vor 7 Monaten

Introducing City Roots... It's more than where we're from. It's who we are. To shop the brand new collection or to learn more about our City Roots series visit Minor League Baseball
2:35

Introducing City Roots... It's more than where we're from. It's who we are. To shop the brand new collection or to learn more about our City Roots series visit Minor League Baseball

Albuquerque Isotopes

21,379 Aufrufe • vor 1 Jahr

Today, we celebrated the groundbreaking for our new School of Medicine in the historic BullStreet District in downtown Columbia, a vision more than a decade in the making. Learn more about the School of Medicine Columbia and our Health Sciences Campus. 🩺
0:52

Today, we celebrated the groundbreaking for our new School of Medicine in the historic BullStreet District in downtown Columbia, a vision more than a decade in the making. Learn more about the School of Medicine Columbia and our Health Sciences Campus. 🩺

University of South Carolina

15,708 Aufrufe • vor 1 Jahr

One of our first Hall of Fame inductees 🇸🇪 Learn more about a true legend of the game Tommy ”Potti” Ingemarsson from our interview at the recent HLTV Awards: 📺
0:30

One of our first Hall of Fame inductees 🇸🇪 Learn more about a true legend of the game Tommy ”Potti” Ingemarsson from our interview at the recent HLTV Awards: 📺

HLTV.org

49,774 Aufrufe • vor 1 Jahr

Today on the 5th Foundation Day of National Democratic Movement we pay tribute to our martyrs. NDM has achieved a lot in a short time with support from our people. We are proud of the resilience of our members. We will continue with our political struggle for rights and justice. #NDM5Years
0:28

Today on the 5th Foundation Day of National Democratic Movement we pay tribute to our martyrs. NDM has achieved a lot in a short time with support from our people. We are proud of the resilience of our members. We will continue with our political struggle for rights and justice. #NDM5Years

Mohsin Dawar

12,784 Aufrufe • vor 25 Tagen

Introducing the cutest Fenders you'll ever see. We teamed up with Loog to make mini versions of our most iconic models — a perfect beginner guitar for kids of all ages. Learn more here:
0:39

Introducing the cutest Fenders you'll ever see. We teamed up with Loog to make mini versions of our most iconic models — a perfect beginner guitar for kids of all ages. Learn more here:

Fender

139,212 Aufrufe • vor 3 Jahren

Sunday sermon... What a speech and address to the nation from a head of state! We need more of these for our continent. Listen to him.
21:36

Sunday sermon... What a speech and address to the nation from a head of state! We need more of these for our continent. Listen to him.

Mzalendo Mūtemi wa Kīama

17,856 Aufrufe • vor 1 Jahr

This #EarthDay, we're reflecting on our sustainability journey and working to build a bright future for our stakeholders, people and planet. Hear from our Dir. of Sustainability to learn more about our approach to circularity:
0:47

This #EarthDay, we're reflecting on our sustainability journey and working to build a bright future for our stakeholders, people and planet. Hear from our Dir. of Sustainability to learn more about our approach to circularity:

SHEIN News

194,570 Aufrufe • vor 3 Jahren

We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.
2:15

We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.

Physical Intelligence

443,831 Aufrufe • vor 6 Monaten

We’re redefining what’s possible with AI. With the release of our latest model, Command A, optimized for real-world agentic and multilingual tasks, we’re demonstrating our commitment to bringing enterprises AI that goes beyond the ordinary, and offers security & efficiency. Our team has developed highly capable and efficient models that can be run on just 2 GPUs. Check out our tech report to learn more:
0:30

We’re redefining what’s possible with AI. With the release of our latest model, Command A, optimized for real-world agentic and multilingual tasks, we’re demonstrating our commitment to bringing enterprises AI that goes beyond the ordinary, and offers security & efficiency. Our team has developed highly capable and efficient models that can be run on just 2 GPUs. Check out our tech report to learn more:

Cohere

16,426 Aufrufe • vor 1 Jahr

Our model can now learn from its own experience with RL! Our new π*0.6 model can more than double throughput over a base model trained without RL, and can perform real-world tasks: making espresso drinks, folding diverse laundry, and assembling boxes. More in the thread below.
3:50

Our model can now learn from its own experience with RL! Our new π*0.6 model can more than double throughput over a base model trained without RL, and can perform real-world tasks: making espresso drinks, folding diverse laundry, and assembling boxes. More in the thread below.

Physical Intelligence

710,612 Aufrufe • vor 10 Monaten

Today, we are excited to release FLUX.1 Tools, a suite of models designed to add control and steerability to our base text-to-image model FLUX.1, enabling the modification and re-creation of real and generated images. Learn more in our blogpost:
0:29

Today, we are excited to release FLUX.1 Tools, a suite of models designed to add control and steerability to our base text-to-image model FLUX.1, enabling the modification and re-creation of real and generated images. Learn more in our blogpost:

Black Forest Labs

350,314 Aufrufe • vor 1 Jahr

How does a PGA Tour pro hire and pay a coach? Learn more from Ben Griffin in our latest episode of Ask The Expert:
0:42

How does a PGA Tour pro hire and pay a coach? Learn more from Ben Griffin in our latest episode of Ask The Expert:

Golf Digest

15,236 Aufrufe • vor 4 Monaten