Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Here is a tutorial on training LLaSA (LLaMA-based TTS) using GRPO to improve prosody, rhythm, and expressiveness in synthesized speech with TRL!

steven

2,576 subscribers

15,488 Aufrufe • vor 8 Monaten •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Text-to-Speech is HERE 🎤 Create audiobooks, mini podcasts, and your own TTS website with Manus!

Text-to-Speech is HERE 🎤 Create audiobooks, mini podcasts, and your own TTS website with Manus!

Manus

198,081 Aufrufe • vor 1 Jahr

New research from Databricks: LLMs Can Learn to Reason via Off-Policy RL Optimal Advantage-based Policy Optimization with Lagged Inference policy (OAPL) shows you don’t need strict on-policy training to improve reasoning. It matches or beats Group Relative Policy Optimization (GRPO), stays stable with large policy lag, and uses ~3× fewer training generations. For Databricks customers, it’s a simpler, practical, and equally powerful approach to RL that Databricks is pioneering internally — and bringing directly to Databricks customers, so enterprises can improve agents using the same methods we use for our in-house agents, without complex infrastructure changes.

New research from Databricks: LLMs Can Learn to Reason via Off-Policy RL Optimal Advantage-based Policy Optimization with Lagged Inference policy (OAPL) shows you don’t need strict on-policy training to improve reasoning. It matches or beats Group Relative Policy Optimization (GRPO), stays stable with large policy lag, and uses ~3× fewer training generations. For Databricks customers, it’s a simpler, practical, and equally powerful approach to RL that Databricks is pioneering internally — and bringing directly to Databricks customers, so enterprises can improve agents using the same methods we use for our in-house agents, without complex infrastructure changes.

Databricks AI Research

12,539 Aufrufe • vor 4 Monaten

🚀 Meet Qwen-TTS – now live via the Qwen API ! Trained on millions of hours of speech, it delivers ultra-natural, expressive audio with smart prosody, pacing, and emotion. 🗣️ Supports 3 Chinese dialects: Beijing, Shanghai, Sichuan 🎙️ 7 bilingual voices: Cherry, Ethan, Chelsie, Serena, Dylan, Jada, Sunny 🔗 Learn more: 👉 Start using it via API:

🚀 Meet Qwen-TTS – now live via the Qwen API ! Trained on millions of hours of speech, it delivers ultra-natural, expressive audio with smart prosody, pacing, and emotion. 🗣️ Supports 3 Chinese dialects: Beijing, Shanghai, Sichuan 🎙️ 7 bilingual voices: Cherry, Ethan, Chelsie, Serena, Dylan, Jada, Sunny 🔗 Learn more: 👉 Start using it via API:

Qwen

144,751 Aufrufe • vor 1 Jahr

Here is a quick tutorial on how to use the new cl_netgraph in CS2

Here is a quick tutorial on how to use the new cl_netgraph in CS2

Tradeit

246,195 Aufrufe • vor 2 Jahren

This is amazing! Someone trained a 1.5B model (with GRPO) to solve mazes! model:

This is amazing! Someone trained a 1.5B model (with GRPO) to solve mazes! model:

Victor M

139,407 Aufrufe • vor 1 Jahr

Tutorial: using GPT Image 2 and Seedance 2 to create beautiful animated children's story. Here is how I created this. 👇

Tutorial: using GPT Image 2 and Seedance 2 to create beautiful animated children's story. Here is how I created this. 👇

Nimentrix

21,148 Aufrufe • vor 2 Monaten

Here's a tutorial on how to take footage and turn it into a fictive character using Kling O1 on Freepik

Here's a tutorial on how to take footage and turn it into a fictive character using Kling O1 on Freepik

Martin LeBlanc

12,540 Aufrufe • vor 7 Monaten

We found a new way to get language models to reason. 🤯 No RL, no training, no verifiers, no prompting. ❌ With better sampling, base models can achieve single-shot reasoning on par with (or better than!) GRPO while avoiding its characteristic loss in generation diversity.

We found a new way to get language models to reason. 🤯 No RL, no training, no verifiers, no prompting. ❌ With better sampling, base models can achieve single-shot reasoning on par with (or better than!) GRPO while avoiding its characteristic loss in generation diversity.

Aayush Karan

277,459 Aufrufe • vor 8 Monaten

Wow! 🎉 MAU personally came to Fansland Web3 Music Festival, bringing unprecedented passion! 🔥🎶 I can't wait to use my cat superpowers to ignite every note and energy here! 🌟 Here is a perfect combination of digital avatar and music. Every node is a note of dreams, and every rhythm is a carnival in the future! 🎧✨ Join us on the memecoin stage and release our crazy rhythm with me! 💥#AIIdol #MAU #AI #memecoin #solanamemecoin

MAU

14,474 Aufrufe • vor 1 Jahr

🚨 Marco Rubio just CRUSHED it. Even liberals don't know what to say. My timeline is flooded with people raving about his speech in Munich that focused on defending the west, Christianity, and opposing mass migration. Based Marco is a force to be reckoned with, that's a FACT!

🚨 Marco Rubio just CRUSHED it. Even liberals don't know what to say. My timeline is flooded with people raving about his speech in Munich that focused on defending the west, Christianity, and opposing mass migration. Based Marco is a force to be reckoned with, that's a FACT!

Eric Daugherty

273,547 Aufrufe • vor 4 Monaten

Tutorial on how to add an embroidered effect on any object in illustrator with no plugin, using object > create mosaic + global edit

Tutorial on how to add an embroidered effect on any object in illustrator with no plugin, using object > create mosaic + global edit

After Effects Tricks!

70,859 Aufrufe • vor 4 Monaten

Using here PolyQuilt addon (free) for topology. Useful smooth relax brush and easy to use. #b3d Tutorial:

Using here PolyQuilt addon (free) for topology. Useful smooth relax brush and easy to use. #b3d Tutorial:

Farrukh Abdur

189,885 Aufrufe • vor 2 Jahren

MiniMax Audio just leveled up with the new Speech-02 model! Turn any file or URL into lifelike audio instantly. Create audiobooks and podcasts effortlessly with up to 200k-character in a single input. Enjoy ultra-realistic TTS in 30+ languages with native flair, unlimited voice cloning, and sub-second streaming! Here are 4 examples:

MiniMax Audio just leveled up with the new Speech-02 model! Turn any file or URL into lifelike audio instantly. Create audiobooks and podcasts effortlessly with up to 200k-character in a single input. Enjoy ultra-realistic TTS in 30+ languages with native flair, unlimited voice cloning, and sub-second streaming! Here are 4 examples:

MiniMax (official)

4,233,495 Aufrufe • vor 1 Jahr

Larry Wheels is TRAINING to IMPROVE his SPEED and FOOTWORK for BOXING 😭👀

Larry Wheels is TRAINING to IMPROVE his SPEED and FOOTWORK for BOXING 😭👀

Cage ☆

203,139 Aufrufe • vor 1 Monat

In case you see a ghost this Halloween Here is a quick tutorial from Seonghwa on how to beat them up👍

In case you see a ghost this Halloween Here is a quick tutorial from Seonghwa on how to beat them up👍

Jazz | Silver&Gold |

44,567 Aufrufe • vor 9 Monaten

Soft is a camera app that captures the rhythm of your environment using the analog sensors in your phone. Here are 100 more beta invites.

Soft is a camera app that captures the rhythm of your environment using the analog sensors in your phone. Here are 100 more beta invites.

Jake Dow-Smith

38,970 Aufrufe • vor 11 Monaten

🆘🆘 HERE IS A TUTORIAL ON HOW TO VOTE ON MNET PLZ EVERYONE VOTE NOWW 🆘🆘

🆘🆘 HERE IS A TUTORIAL ON HOW TO VOTE ON MNET PLZ EVERYONE VOTE NOWW 🆘🆘

💛🐣𝒜𝐵𝐵𝐼𝐸 🐣💛

62,222 Aufrufe • vor 1 Jahr

Better text-to-speech is possible, but none of AWS, Google, Microsoft, etc have released upgrades to their voices in years. We're excited that Unreal Speech has filled this void and we'll be switching voices for Reader's TTS feature soon. Check out a sample 🎼

Better text-to-speech is possible, but none of AWS, Google, Microsoft, etc have released upgrades to their voices in years. We're excited that Unreal Speech has filled this void and we'll be switching voices for Reader's TTS feature soon. Check out a sample 🎼

Reader

70,028 Aufrufe • vor 2 Jahren

One-shot with Claude Fable 5, and this subscription plan UI is live (based on my design using Figma MCP).

One-shot with Claude Fable 5, and this subscription plan UI is live (based on my design using Figma MCP).

Maxim Kuznetsov

13,573 Aufrufe • vor 24 Tagen