正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Introducing Octave 2: our next-generation multilingual text-to-speech model What’s new: - Fluent in 11+ languages - 40% faster (<200ms latency⁠⁠) & 50% cheaper than Octave 1 - Multi-speaker conversation - More reliable pronunciation - New voice conversion & phoneme editing capabilities For the month of October, we’re offering 50%... show more

Hume AI

23,213 subscribers

7,077,598 次观看 • 11 个月前 •via X (Twitter)

健康养生科学技术教育

Anya Rossi• Live Now

Private livecam show

48 条评论

Hume AI 的头像

Hume AI11 个月前

Our new models speak over 11 languages, including English, Japanese 🇯🇵, Korean 🇰🇷, Spanish 🇪🇸, French 🇫🇷, Portuguese 🇵🇹, Italian 🇮🇹, German 🇩🇪, Russian 🇷🇺, Hindi 🇮🇳, and Arabic 🇦🇪. They’re faster (<200ms), more efficient, and more accurate. We are also rolling out a conversationally tuned model, EVI 4 mini, that is powered by the intelligence of any frontier model like Claude 4.5 Sonnet or GPT-5. Try our live translation demo & EVI 4 mini chat, built on a few voice samples and a prompt:

Hume AI 的头像

Hume AI11 个月前

We'll be releasing more evaluations, more languages, and access to voice conversion and phoneme editing soon. In the meantime, we’re excited to see what you create! Read more about product & research in our blog ↓

Hume AI 的头像

Hume AI11 个月前

We’re also HIRING across roles! Apply below on our careers page

Umesh 的头像

Umesh11 个月前

Congratulations 👏 It's amazing!

Hume AI 的头像

Hume AI11 个月前

thanks for the support umesh!

proper 的头像

proper11 个月前

2.0!! congrats on the launch!

Kol Tregaskes 的头像

Kol Tregaskes11 个月前

Nice guys, thanks for the early access!

Hume AI 的头像

Hume AI11 个月前

would love to hear what you think Kol!

Alan Cowen 的头像

Alan Cowen11 个月前

日本語に最適な音声AI

Hume AI 的头像

Hume AI11 个月前

The land of @BIG_CLAPPY!!

mori 的头像

mori11 个月前

LFG!! ⭐️

mori 的头像

mori11 个月前

can’t wait to see what ppl build w thisss

Hume AI 的头像

Hume AI11 个月前

show us what you build!

Alec Freudenstein 的头像

Alec Freudenstein11 个月前

INCREDIBLE WORK TEAM!!!! SO PROUD OF EVERYONE WHO PUT SO MUCH TIME INTO THIS!!!! NOBODY SHIPS LIKE HUME!!!!!!!!

Janet Ho 的头像

Janet Ho11 个月前

Can’t wait to go to Japan with this

Linus ✦ Ekenstam 的头像

Linus ✦ Ekenstam11 个月前

Guys, thanks for giving me early access to this, been sharing with the family last couple of days, and it’s been a blast. Extremely bullish on the trajectory you are on, keep shipping, this upgrade is spot on. 🫡

Hume AI 的头像

Hume AI11 个月前

thanks Linus!!

Lucas | IA 的头像

Lucas | IA11 个月前

SO good

Hume AI 的头像

Hume AI11 个月前

thank you 😊

Alvaro Cintas 的头像

Alvaro Cintas11 个月前

SO good!

Miguel Ángel | GptZone 的头像

Miguel Ángel | GptZone11 个月前

I am just trying it right now and it seems spectacular!!

Robert Scoble 的头像

Robert Scoble11 个月前

Great getting to talk to your CEO about this launch:

Hume AI 的头像

Hume AI11 个月前

awesome conversation!

Abdul Shakoor 的头像

Abdul Shakoor11 个月前

octave 2 feels like a serious upgrade.

Hume AI 的头像

Hume AI11 个月前

so glad you think so!

cate bligh 的头像

cate bligh11 个月前

Great video! I'm looking forward to playing with this more!

Hume AI 的头像

Hume AI11 个月前

love the videos you create with us cate!

Dr. Apolotary 的头像

Dr. Apolotary11 个月前

Why is he talking with a strong English accent in Japanese

Hume AI 的头像

Hume AI11 个月前

because he’s English! our model preserves accents based on the language of the base voice

Adam 的头像

Adam11 个月前

Unbelievable! I'll talk about it soon 👀

Jack 的头像

Jack11 个月前

So amazing

Nelly; 的头像

Nelly;11 个月前

no excuses now.. I can actually sound fluent in 11 languages😂

Qing Liu 的头像

Qing Liu11 个月前

Soooo proud 🥺 This is really amazing!!

@Czr305 的头像

@Czr30511 个月前

What about for EVI?

Hume AI 的头像

Hume AI11 个月前

we have EVI4 mini also in preview

@Czr305 的头像

@Czr30511 个月前

How do I gain access ?

Eleftheria Batsou 的头像

Eleftheria Batsou11 个月前

Let me check that!!! Excited for all these features

ayn 的头像

ayn11 个月前

Congratulations HUME!! 🫶🏻

cova 的头像

cova11 个月前

Congratulations!! 🫶

WISe Social 的头像

WISe Social11 个月前

Exciting advancements with Octave 2. Its multi-speaker feature could really enhance AI agent interactions. Imagine seamless transactions through secure channels using satellite connectivity, perfect for a world where speed and reliability are key.

Luke Miler 的头像

Luke Miler11 个月前

sweet! will add to shortcut

Afiz ⚡️ 的头像

Afiz ⚡️11 个月前

multilingual text-to-speech. I would definitely want to give it a try.

Jameson McSmirnoff 🌊🛥 的头像

Jameson McSmirnoff 🌊🛥11 个月前

Deutsch ist euer Endgegner!

Roni Rahman 的头像

Roni Rahman11 个月前

Looks really good!

MetaDJ 的头像

MetaDJ11 个月前

Awesome! 🔥

Jussi 的头像

Jussi11 个月前

Will it support audio to audio? I am looking for a audio model to change the vocals of a song I generated on Suno to my own voice

Jaden Tripp 的头像

Jaden Tripp11 个月前

IPA input?

Rafael Estrela | IA 的头像

Rafael Estrela | IA11 个月前

Incrivel !

相关视频

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:

Stability AI

976,927 次观看 • 2 年前

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.

BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2. > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article) > MAI Transcribe 1.5, a new model for speech-to-text use cases. > MAI Image 2.5, already announced last week, is now available on LM Arena in preview. Compared to MAI Image 2, it supports file uploads and can be used for image editing.

🚨 AI News | TestingCatalog

46,535 次观看 • 3 个月前

Introducing GPT-Realtime-2 in the API: our most intelligent voice model yet, bringing GPT-5-class reasoning to voice agents. Voice agents are now real-time collaborators that can listen, reason, and solve complex problems as conversations unfold. Now available in the API alongside streaming models GPT-Realtime-Translate and GPT-Realtime-Whisper — a new set of audio capabilities for the next generation of voice interfaces.

Introducing GPT-Realtime-2 in the API: our most intelligent voice model yet, bringing GPT-5-class reasoning to voice agents. Voice agents are now real-time collaborators that can listen, reason, and solve complex problems as conversations unfold. Now available in the API alongside streaming models GPT-Realtime-Translate and GPT-Realtime-Whisper — a new set of audio capabilities for the next generation of voice interfaces.

OpenAI

3,664,918 次观看 • 4 个月前

QUANT TEKEL LAUNCH - 50% OFF Introducing Quant Tekel: The Future Of Trading! Formerly AscendX Capital, we’re now broker owned to bring you an even more powerful trading experience. To Celebrate our launch we’re offering 50% off all Evaluations! • Broker Owned: Built with sustainability in mind. • 1, 2, 3 Step Challenges: Choose your path to success. • ⁠90% Profit Split: Keep what you earn. • ⁠Over $6,000,000 in Trader Payouts: We reward your skills. • ⁠14 Day Payouts: Fast and reliable. 🌐 Join us and take your trading to the next level! #QuantTekel #TheFutureOfTrading

QUANT TEKEL LAUNCH - 50% OFF Introducing Quant Tekel: The Future Of Trading! Formerly AscendX Capital, we’re now broker owned to bring you an even more powerful trading experience. To Celebrate our launch we’re offering 50% off all Evaluations! • Broker Owned: Built with sustainability in mind. • 1, 2, 3 Step Challenges: Choose your path to success. • ⁠90% Profit Split: Keep what you earn. • ⁠Over $6,000,000 in Trader Payouts: We reward your skills. • ⁠14 Day Payouts: Fast and reliable. 🌐 Join us and take your trading to the next level! #QuantTekel #TheFutureOfTrading

QT FUNDED

22,499 次观看 • 1 年前

Introducing Soniox TTS v2, our most powerful text-to-speech model yet. Soniox TTS v2 brings extraordinary voice quality, expressive control through audio tags, exceptional precision, high-fidelity voice cloning, more than 60 languages, natural language mixing, and low-latency streaming together in one model. Built from the ground up with a new model architecture, audio codec, and highly optimized inference engine, Soniox TTS v2 delivers frontier TTS at just $0.70 per generated hour. Available globally today as tts-rt-v2. Try it out: Turn the sound on and hear what’s possible. 🔊

Introducing Soniox TTS v2, our most powerful text-to-speech model yet. Soniox TTS v2 brings extraordinary voice quality, expressive control through audio tags, exceptional precision, high-fidelity voice cloning, more than 60 languages, natural language mixing, and low-latency streaming together in one model. Built from the ground up with a new model architecture, audio codec, and highly optimized inference engine, Soniox TTS v2 delivers frontier TTS at just $0.70 per generated hour. Available globally today as tts-rt-v2. Try it out: Turn the sound on and hear what’s possible. 🔊

Soniox

663,421 次观看 • 1 个月前

We partnered with artists, designers, and builders to create new AI tools that solve real problems in their creative workflows. Here’s what’s new: — Introducing Google Pics in Google Workspace: A brand-new image creation & editing tool. Move and resize objects, add text, and translate just by hovering and clicking — Big updates to صافي النيه😉: 1) You can now create with Gemini Omni Flash in Google Flow 2) Google Flow Agent is a multi-step creative partner that reasons and plans complex tasks with you. 3) Google Flow tools are custom tools you can “vibe code” for animations, video effects, text layering & more — Design live with Stitch by Google: Now, you can use text or voice prompts to edit layouts in real time then export those designs straight to code — More creative control in صافي النيه😉Music: Edit songs section by section, remix the style of full songs, and create music videos with our new Gemini Omni Flash model

Google AI

13,954,152 次观看 • 3 个月前

Today we’re announcing r-1, our new document parsing model. It’s more accurate than our most powerful agentic OCR models, faster, and up to 6x cheaper. At Reducto, we spent two years building specialized models for complex visual layouts, tables spanning multiple pages, and key formatting like strikethroughs. We then used everything we learned to build r-1, the first in a new generation of models designed to handle the hardest documents without multiplying cost. This early preview delivers a 20% lower error rate than our most accurate legacy agentic models and will keep improving with new checkpoints over the next few weeks. Accuracy is only the beginning. r-1 preview is available at 1¢ per page all-in, with additional volume discounts as you scale. In the near term we’re also going to release r-1 mini, and an auto mode that intelligently selects the right approach for each page. If you’re currently using another parser, we’re offering up to $5,000 in credits to evaluate and migrate. You can claim the migration offer and learn more about r-1 using the links in the comments. Happy parsing!

Today we’re announcing r-1, our new document parsing model. It’s more accurate than our most powerful agentic OCR models, faster, and up to 6x cheaper. At Reducto, we spent two years building specialized models for complex visual layouts, tables spanning multiple pages, and key formatting like strikethroughs. We then used everything we learned to build r-1, the first in a new generation of models designed to handle the hardest documents without multiplying cost. This early preview delivers a 20% lower error rate than our most accurate legacy agentic models and will keep improving with new checkpoints over the next few weeks. Accuracy is only the beginning. r-1 preview is available at 1¢ per page all-in, with additional volume discounts as you scale. In the near term we’re also going to release r-1 mini, and an auto mode that intelligently selects the right approach for each page. If you’re currently using another parser, we’re offering up to $5,000 in credits to evaluate and migrate. You can claim the migration offer and learn more about r-1 using the links in the comments. Happy parsing!

Adit

298,438 次观看 • 13 天前

Today, we’re introducing new tools and model updates to help you build, deploy, and scale Voice AI applications.🎙️ 🆕 Speech Understanding: Turn transcripts into actionable data with speaker identification, custom formatting & translation 🆕 LLM Gateway: One API for your voice-to-intelligence pipeline, with GPT, Claude, Gemini & more 🆕 Voice AI Guardrails: End-to-end protection for safe, compliant content Model Upgrades: ✅ 99 languages with auto code-switching ✅ 64% fewer speaker counting errors ✅ 57% better accuracy on critical terms with 1,000-word context Real results: 🔥 Calabrio: 80% boost in customer satisfaction, 22% revenue increase 🔥 Siro: "10/10 customers say 'wow, that insight was crisp'" Build & scale Voice AI in minutes! Try it in our Playground or check the docs. 🚀

Today, we’re introducing new tools and model updates to help you build, deploy, and scale Voice AI applications.🎙️ 🆕 Speech Understanding: Turn transcripts into actionable data with speaker identification, custom formatting & translation 🆕 LLM Gateway: One API for your voice-to-intelligence pipeline, with GPT, Claude, Gemini & more 🆕 Voice AI Guardrails: End-to-end protection for safe, compliant content Model Upgrades: ✅ 99 languages with auto code-switching ✅ 64% fewer speaker counting errors ✅ 57% better accuracy on critical terms with 1,000-word context Real results: 🔥 Calabrio: 80% boost in customer satisfaction, 22% revenue increase 🔥 Siro: "10/10 customers say 'wow, that insight was crisp'" Build & scale Voice AI in minutes! Try it in our Playground or check the docs. 🚀

AssemblyAI

18,157 次观看 • 10 个月前

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

Tencent Hy

89,392 次观看 • 1 年前

$LockVaultEV IS OFFICIALLY HERE 💻 LockVault has officially launched in Beta! As a thank you for joining during our beta phase, we’re offering 50% OFF recurring for all beta members for LIFE. Use code: “BETA” at checkout to lock in your discount. ⚠️ Limited spots available — this offer is only for early beta users only the first 50 members can use this code. Don’t miss out on the sharpest betting intelligence tool on the market. Lock The Edge Vault The Profit! 🔒 What our tool offers Player Prop History DFS Optimizer Arbitrage Scanner Game Lines Optimizer Get all of your info from one place for a fraction of the price 🤝 Join here ⬇️$

LockVaultEV IS OFFICIALLY HERE 💻 LockVault has officially launched in Beta! As a thank you for joining during our beta phase, we’re offering 50% OFF recurring for all beta members for LIFE. Use code: “BETA” at checkout to lock in your discount. ⚠️ Limited spots available — this offer is only for early beta users only the first 50 members can use this code. Don’t miss out on the sharpest betting intelligence tool on the market. Lock The Edge Vault The Profit! 🔒 What our tool offers Player Prop History DFS Optimizer Arbitrage Scanner Game Lines Optimizer Get all of your info from one place for a fraction of the price 🤝 Join here ⬇️

Angelo

155,426 次观看 • 3 个月前

Better/faster/cheaper voice AI turn detection with Gemini 2.0 The code that determines when the agent should respond to the user is some of the most important code in your voice AI agent. The technical terms for this job are "turn detection" or "phrase endpointing." If the voice AI responds before the user has finished their thought, the conversation is choppy and unproductive. If the AI waits too long, the conversation is slow and frustrating. There are a number of ways to approach this. You can: 1. Use a fast "voice activity detection" model to detect pauses in speech. Respond when the user pauses. 2. Use a specialized phrase endpointing model that operates on transcribed text, pattern-matching on text semantics. 3. Train a specialized phrase endpointing model that operates directly on audio. 4. Leverage the native audio capabilities of a SOTA LLM like Gemini 2.0. We've benchmarked all four of these, and Gemini 2.0 currently beats other approaches. Using Gemini is also cheaper than transcribing the audio separately using a transcription service or model. Here's a short video showing Gemini phrase endpointing in two scenarios. First, correctly handling pauses in natural conversation. Second, requesting a phone number (which is a common activity in a use case like customer support). You can see the Completeness check lines in the terminal output, printed each time Gemini processes a chunk of audio.

Better/faster/cheaper voice AI turn detection with Gemini 2.0 The code that determines when the agent should respond to the user is some of the most important code in your voice AI agent. The technical terms for this job are "turn detection" or "phrase endpointing." If the voice AI responds before the user has finished their thought, the conversation is choppy and unproductive. If the AI waits too long, the conversation is slow and frustrating. There are a number of ways to approach this. You can: 1. Use a fast "voice activity detection" model to detect pauses in speech. Respond when the user pauses. 2. Use a specialized phrase endpointing model that operates on transcribed text, pattern-matching on text semantics. 3. Train a specialized phrase endpointing model that operates directly on audio. 4. Leverage the native audio capabilities of a SOTA LLM like Gemini 2.0. We've benchmarked all four of these, and Gemini 2.0 currently beats other approaches. Using Gemini is also cheaper than transcribing the audio separately using a transcription service or model. Here's a short video showing Gemini phrase endpointing in two scenarios. First, correctly handling pauses in natural conversation. Second, requesting a phone number (which is a common activity in a use case like customer support). You can see the Completeness check lines in the terminal output, printed each time Gemini processes a chunk of audio.

kwindla

64,190 次观看 • 1 年前

Last month, we released Lyria 3, enabling you to create tracks with lyrics from text, image, or video prompts. Now, we’re introducing Lyria 3 Pro, which expands upon our music generation model to offer additional advanced capabilities. What’s really special about this upgrade is that the model now understands the architecture of music. This makes it possible to prompt for intros, verses, choruses and bridges + generate songs with more complex transitions. You can also create tracks up to 3 minutes long, a big change from previous models that were limited to 30 second tracks. Use Lyria 3 Pro to build upon your existing creativity. We’re excited for your beats to drop 🎶

Last month, we released Lyria 3, enabling you to create tracks with lyrics from text, image, or video prompts. Now, we’re introducing Lyria 3 Pro, which expands upon our music generation model to offer additional advanced capabilities. What’s really special about this upgrade is that the model now understands the architecture of music. This makes it possible to prompt for intros, verses, choruses and bridges + generate songs with more complex transitions. You can also create tracks up to 3 minutes long, a big change from previous models that were limited to 30 second tracks. Use Lyria 3 Pro to build upon your existing creativity. We’re excited for your beats to drop 🎶

Google AI

81,583 次观看 • 5 个月前

Voice used to be AI’s forgotten modality - now it's having its big moment: rapid innovation, big funding rounds, major agentic applications My conversation with Neil Zeghidour, top AI researcher in the field (Google DeepMind, Meta, kyutai) and now CEO of Gradium This is a reference episode on all things voice AI 🔥 00:00 Intro 01:21 Voice AI’s big moment, and why we’re still early 03:34 Why voice lagged behind text/image/video 06:06 The convergence era: transformers for every modality 07:40 Beyond Her: always-on assistants, wake words, voice-first devices 11:01 Voice vs text: where voice fits (even for coding) 12:56 Neil’s origin story: from finance to machine learning, with help from Yann LeCun and Soumith Chintala 18:35 Neural codecs (SoundStream): compression as the unlock 22:30 Kyutai: open research, small elite teams, moving fast 31:32 Why big labs haven’t “won” voice AI4 34:01 On-device voice: where it works, why compact models matter 46:37 The last mile: real-world robustness, pronunciation, uptime 41:35 Benchmarking voice: why metrics fail, how they actually test 47:03 Cascades vs speech-to-speech: trade-offs + what’s next 54:05 Hardest frontier: noisy rooms, factories, multi-speaker chaos 1:00:50 New languages + dialects: what transfers, what doesn’t 1:02:54 Hardware & compute: why voice isn’t a 10,000-GPU game 1:07:27 What data do you need to train voice models 1:09:02 Deepfakes + privacy: why watermarking isn’t a solution 1:12:30 Voice + vision: multimodality, screen awareness, video+audio 1:14:43 Voice cloning vs voice design: where the market goes 1:16:32 Paris/Europe AI: talent density, underdog energy, what’s next

Voice used to be AI’s forgotten modality - now it's having its big moment: rapid innovation, big funding rounds, major agentic applications My conversation with Neil Zeghidour, top AI researcher in the field (Google DeepMind, Meta, kyutai) and now CEO of Gradium This is a reference episode on all things voice AI 🔥 00:00 Intro 01:21 Voice AI’s big moment, and why we’re still early 03:34 Why voice lagged behind text/image/video 06:06 The convergence era: transformers for every modality 07:40 Beyond Her: always-on assistants, wake words, voice-first devices 11:01 Voice vs text: where voice fits (even for coding) 12:56 Neil’s origin story: from finance to machine learning, with help from Yann LeCun and Soumith Chintala 18:35 Neural codecs (SoundStream): compression as the unlock 22:30 Kyutai: open research, small elite teams, moving fast 31:32 Why big labs haven’t “won” voice AI4 34:01 On-device voice: where it works, why compact models matter 46:37 The last mile: real-world robustness, pronunciation, uptime 41:35 Benchmarking voice: why metrics fail, how they actually test 47:03 Cascades vs speech-to-speech: trade-offs + what’s next 54:05 Hardest frontier: noisy rooms, factories, multi-speaker chaos 1:00:50 New languages + dialects: what transfers, what doesn’t 1:02:54 Hardware & compute: why voice isn’t a 10,000-GPU game 1:07:27 What data do you need to train voice models 1:09:02 Deepfakes + privacy: why watermarking isn’t a solution 1:12:30 Voice + vision: multimodality, screen awareness, video+audio 1:14:43 Voice cloning vs voice design: where the market goes 1:16:32 Paris/Europe AI: talent density, underdog energy, what’s next

Matt Turck

22,980 次观看 • 6 个月前

We’re still buzzing from my running mate Nicole Shanahan’s amazing speech in Oakland last month and the new energy she’s brought to our campaign. We’re here to radically change the political conversation and to put issues on the map that both parties have neglected. While about one in 5 voters already supports us, to get on the ballot in all 50 states and the District of Columbia and take the White House, we need financial support. The April 30th FEC deadline is around the corner, and your contribution today means more than ever. DONATE NOW ➡️ #KennedyShanahan24

We’re still buzzing from my running mate Nicole Shanahan’s amazing speech in Oakland last month and the new energy she’s brought to our campaign. We’re here to radically change the political conversation and to put issues on the map that both parties have neglected. While about one in 5 voters already supports us, to get on the ballot in all 50 states and the District of Columbia and take the White House, we need financial support. The April 30th FEC deadline is around the corner, and your contribution today means more than ever. DONATE NOW ➡️ #KennedyShanahan24

Robert F. Kennedy Jr

257,170 次观看 • 2 年前

Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new Llama 3.1 models including our long awaited 405B. These models deliver improved reasoning capabilities, a larger 128K token context window and improved support for 8 languages among other improvements. Llama 3.1 405B rivals leading closed source models on state-of-the-art capabilities across a range of tasks in general knowledge, steerability, math, tool use and multilingual translation. The models are available to download now directly from Meta or Hugging Face. With today’s release the ecosystem is also ready to go with 25+ partners rolling out our latest models — including Amazon Web Services, NVIDIA, Databricks, GroqInc, Dell Technologies, Microsoft Azure and Google Cloud ready on day one. More details in the full announcement ➡️ Download Llama 3.1 models ➡️ With these releases we’re setting the stage for unprecedented new opportunities and we can’t wait to see the innovation our newest models will unlock across all levels of the AI community.

Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new Llama 3.1 models including our long awaited 405B. These models deliver improved reasoning capabilities, a larger 128K token context window and improved support for 8 languages among other improvements. Llama 3.1 405B rivals leading closed source models on state-of-the-art capabilities across a range of tasks in general knowledge, steerability, math, tool use and multilingual translation. The models are available to download now directly from Meta or Hugging Face. With today’s release the ecosystem is also ready to go with 25+ partners rolling out our latest models — including Amazon Web Services, NVIDIA, Databricks, GroqInc, Dell Technologies, Microsoft Azure and Google Cloud ready on day one. More details in the full announcement ➡️ Download Llama 3.1 models ➡️ With these releases we’re setting the stage for unprecedented new opportunities and we can’t wait to see the innovation our newest models will unlock across all levels of the AI community.

AI at Meta

1,269,771 次观看 • 2 年前

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

AI at Meta

150,477 次观看 • 1 年前

Carney: Now, above all, we will build Canadian. The new security threats reaching our shores mean we’re making the biggest investments in our military in generations. But up until now, more than 70 cents of every dollar of military capital spending by our government has gone to the United States. With our new Defense Investment Agency, we’ll make sure we build more of our security and defense capabilities here at home—creating new careers in aerospace, shipbuilding, cyber, and AI for Canadian engineers, technicians, and scientists. It’s the same thing with steel. Consider that while 80% of the steel Europe buys is manufactured in Europe, and 75% of the steel the U.S. buys is made in the U.S., only 40% of the steel we use is made here. That doesn’t make sense—not when we have the people, the resources, and the companies to build right here in Canada. So, our upcoming budget will move forward with our Buy Canadian policy—prioritizing Canadian steel, aluminum, lumber, manufactured goods, and technology for Canadian projects. With new orders for homes, infrastructure, and defense equipment becoming orders for Canadian suppliers, we will be our own best customer

Carney: Now, above all, we will build Canadian. The new security threats reaching our shores mean we’re making the biggest investments in our military in generations. But up until now, more than 70 cents of every dollar of military capital spending by our government has gone to the United States. With our new Defense Investment Agency, we’ll make sure we build more of our security and defense capabilities here at home—creating new careers in aerospace, shipbuilding, cyber, and AI for Canadian engineers, technicians, and scientists. It’s the same thing with steel. Consider that while 80% of the steel Europe buys is manufactured in Europe, and 75% of the steel the U.S. buys is made in the U.S., only 40% of the steel we use is made here. That doesn’t make sense—not when we have the people, the resources, and the companies to build right here in Canada. So, our upcoming budget will move forward with our Buy Canadian policy—prioritizing Canadian steel, aluminum, lumber, manufactured goods, and technology for Canadian projects. With new orders for homes, infrastructure, and defense equipment becoming orders for Canadian suppliers, we will be our own best customer

Acyn

239,493 次观看 • 10 个月前

🚀 Next Phase Unlocked: Remittix Bonus Update Following the incredible momentum that got us past the soft cap, we’re excited to keep things moving with a brand new limited-time offer. As we set our sights on the 20 million dollar milestone, we’re introducing a 40% bonus on all token purchases. This offer is available when you use the promo code WALLET40 during your transaction. Also, our 10% referral bonus remains active, giving you the chance to earn even more by bringing others into the Remittix ecosystem. With tokens still priced at $0.0895, this is another strong opportunity for new and returning investors to get in before the next price adjustment. Let’s keep the momentum going. The journey is far from over. Warm Regards, Remittix Team

Remittix

10,407 次观看 • 1 年前