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。

Just released on Hugging Face: Vui, a 100M open-source NotebookLM! 3 models: > Vui.BASE is the base checkpoint trained on 40k hours of audio conversations > Vui.ABRAHAM is a single speaker model that can reply with context awareness. > Vui.COHOST is checkpoint with two speakers that can talk to... show more

steven

2,545 subscribers

43,620 次观看 • 1 年前 •via X (Twitter)

科学技术教育

Anya Rossi• Live Now

Private livecam show

6 条评论

steven 的头像

steven1 年前

model here: code: here demo here:

HUDI 的头像

HUDI1 年前

🚀 Breaking News! HUDI Datamask Now Integrated with BNB Chain's Greenfield! 🚀 Securely back up your data wallet on decentralized storage—say goodbye to centralized servers! 🔐💾 For the first time in Web3 history, HUDI Datamask empowers you to own your data with the unmatched security of @BNBCHAIN's Greenfield. Experience true data sovereignty and privacy like never before! 🌐✨ Join the revolution in data ownership. Take control today! Learn more 👉 #HUDI #BNBChain #Greenfield #OwnYourData #Web3 #DataSovereignty

Harry Coultas Blum 的头像

Harry Coultas Blum1 年前

Creator here! You can also try the models at

Yossi Dahan 的头像

Yossi Dahan1 年前

The current emphasis seems to be on TTS, but I believe noise cancellation for real-world calls and semantic VAD are far more critical and needed at this stage than another TTS solution.

kfant 的头像

kfant1 年前

sounds good as fast as kokoro?

GTechne 的头像

GTechne1 年前

Vui.COHOST just redefined uncanny valley. Those non-speech tics are genius. How’s the latency on Abraham’s context replies?

相关视频

So you can clone any voice 100% locally using this new open source model?! Alibaba has released Qwen3-TTS on Hugging Face. You can easily: - Create custom voices - Clone any voice from a VERY short audio - Generate speech with style instructions Only 0.6B & 1.8B! Sound on🔊

So you can clone any voice 100% locally using this new open source model?! Alibaba has released Qwen3-TTS on Hugging Face. You can easily: - Create custom voices - Clone any voice from a VERY short audio - Generate speech with style instructions Only 0.6B & 1.8B! Sound on🔊

Paul Couvert

148,975 次观看 • 5 个月前

Releasing vui an open source voice mode 300M TTS model Runs on a single consumer gpu / apple sillicon Context aware speech 6 minutes of context

Releasing vui an open source voice mode 300M TTS model Runs on a single consumer gpu / apple sillicon Context aware speech 6 minutes of context

Harry Coultas Blum

10,865 次观看 • 1 个月前

Krea 2 Open-Source is now available on fal The most aesthetic open-source image model, released with open weights Turbo delivers fast, polished text-to-image at low step counts RAW serves as the malleable base checkpoint for LoRA training

Krea 2 Open-Source is now available on fal The most aesthetic open-source image model, released with open weights Turbo delivers fast, polished text-to-image at low step counts RAW serves as the malleable base checkpoint for LoRA training

fal

46,998 次观看 • 20 天前

Kyutai TTS and Unmute are now open source! The text-to-speech is natural, customizable, and fast: it can serve 32 users with a 350ms latency on a single L40S. Try it out and get started on the project page:

Kyutai TTS and Unmute are now open source! The text-to-speech is natural, customizable, and fast: it can serve 32 users with a 350ms latency on a single L40S. Try it out and get started on the project page:

kyutai

171,579 次观看 • 1 年前

NVIDIA just released a new open source transcription model, Nemotron Speech ASR, designed from the ground up for low-latency use cases like voice agents. Here's a voice agent built with this new model. 24ms transcription finalization and total voice-to-voice inference time under 500ms. This agent actually uses *three* NVIDIA open source models: - Nemotron Speech ASR - Nemotron 3 Nano 30GB in a 4-bit quant (released in December) - A preview checkpoint of the upcoming Magpie text-to-speech model These models are all truly open source: weights, training data, training code, and inference code. This is a big deal! Jensen said in the CES keynote yesterday that he expects open source models to catch up to proprietary models this year in a number of categories. NVIDIA is putting their weight behind making this happen. (As Alan Kay said, the best way to predict the future is to invent it.) The code for this agent is open source too, of course. You can deploy it to production with Modal and Pipecat AI cloud, or run locally on an NVIDIA DGX Spark or RTX 5090.

NVIDIA just released a new open source transcription model, Nemotron Speech ASR, designed from the ground up for low-latency use cases like voice agents. Here's a voice agent built with this new model. 24ms transcription finalization and total voice-to-voice inference time under 500ms. This agent actually uses three NVIDIA open source models: - Nemotron Speech ASR - Nemotron 3 Nano 30GB in a 4-bit quant (released in December) - A preview checkpoint of the upcoming Magpie text-to-speech model These models are all truly open source: weights, training data, training code, and inference code. This is a big deal! Jensen said in the CES keynote yesterday that he expects open source models to catch up to proprietary models this year in a number of categories. NVIDIA is putting their weight behind making this happen. (As Alan Kay said, the best way to predict the future is to invent it.) The code for this agent is open source too, of course. You can deploy it to production with Modal and Pipecat AI cloud, or run locally on an NVIDIA DGX Spark or RTX 5090.

kwindla

274,403 次观看 • 6 个月前

Open Source NotebookLM Alternative on Your PC Wow Dia is REALLY good. 1. Voice clone 2. Non-verbal generation: (laughs), (coughs), .. 3. Dialogues: Just upload 1 clip with 2 voices to generate a NotebookLM style dialogue 1-Click Gradio Launcher. Mac, Windows, Linux.

Open Source NotebookLM Alternative on Your PC Wow Dia is REALLY good. 1. Voice clone 2. Non-verbal generation: (laughs), (coughs), .. 3. Dialogues: Just upload 1 clip with 2 voices to generate a NotebookLM style dialogue 1-Click Gradio Launcher. Mac, Windows, Linux.

cocktail peanut

118,830 次观看 • 1 年前

Open-source audio scene is quite on 🔥 lately! - kyutai STT, TTS modules and Unmute fully open-sourced - NVIDIA drops 3 models: Parakeet (beats Whisper), Audio Flamingo 3 and Canary-Qwen-2.5B (new SOTA on Hugging Face leaderboard) - Mistral AI released 3B and 24B Voxtral

Open-source audio scene is quite on 🔥 lately! - kyutai STT, TTS modules and Unmute fully open-sourced - NVIDIA drops 3 models: Parakeet (beats Whisper), Audio Flamingo 3 and Canary-Qwen-2.5B (new SOTA on Hugging Face leaderboard) - Mistral AI released 3B and 24B Voxtral

Niels Rogge

27,234 次观看 • 1 年前

Microsoft did it again! Speech AI models have a major limitation. They slice long recordings into tiny chunks, lose track of who's speaking, and forget all context halfway through. This is exactly what Microsoft's VibeVoice solves. It's an open-source family of frontier voice AI models for both speech recognition and speech generation. Here's what it can do: > VibeVoice-ASR processes up to 60 minutes of audio in a single pass. No chunking. It outputs structured transcriptions with who spoke, when they spoke, and what they said. > You can feed it custom hotwords like names, technical jargon, or domain-specific terms. The model uses them to significantly improve accuracy on specialized content. > VibeVoice-TTS generates up to 90 minutes of multi-speaker speech with up to 4 distinct speakers. Natural turn-taking, emotional expression, all in one pass. > VibeVoice-Realtime is a 0.5B streaming TTS model with ~300ms first-audio latency. Small enough to deploy practically anywhere. All of this is powered by continuous speech tokenizers running at just 7.5 Hz. This ultra-low frame rate preserves audio quality while making long sequences computationally feasible. I have shared the link to the GitHub repo in the replies!

Microsoft did it again! Speech AI models have a major limitation. They slice long recordings into tiny chunks, lose track of who's speaking, and forget all context halfway through. This is exactly what Microsoft's VibeVoice solves. It's an open-source family of frontier voice AI models for both speech recognition and speech generation. Here's what it can do: > VibeVoice-ASR processes up to 60 minutes of audio in a single pass. No chunking. It outputs structured transcriptions with who spoke, when they spoke, and what they said. > You can feed it custom hotwords like names, technical jargon, or domain-specific terms. The model uses them to significantly improve accuracy on specialized content. > VibeVoice-TTS generates up to 90 minutes of multi-speaker speech with up to 4 distinct speakers. Natural turn-taking, emotional expression, all in one pass. > VibeVoice-Realtime is a 0.5B streaming TTS model with ~300ms first-audio latency. Small enough to deploy practically anywhere. All of this is powered by continuous speech tokenizers running at just 7.5 Hz. This ultra-low frame rate preserves audio quality while making long sequences computationally feasible. I have shared the link to the GitHub repo in the replies!

Akshay 🚀

45,206 次观看 • 3 个月前

NVIDIA just removed one of the biggest friction points in Voice AI. PersonaPlex-7B is an open-source, full-duplex conversational model. Free, open source (MIT), with open model weights on Hugging Face 🤗 Links to repo and weights in 🧵↓ The traditional ASR → LLM → TTS pipeline forces rigid turn-taking. It’s efficient, but it never feels natural. PersonaPlex-7B changes that. This NVIDIA model can listen and speak at the same time. It runs directly on continuous audio tokens with a dual-stream transformer, generating text and audio in parallel instead of passing control between components. That unlocks: → instant back-channel responses → interruptions that feel human → real conversational rhythm Persona control is fully zero-shot! If you’re building low-latency assistants or support agents, this is a big step forward 🔥

NVIDIA just removed one of the biggest friction points in Voice AI. PersonaPlex-7B is an open-source, full-duplex conversational model. Free, open source (MIT), with open model weights on Hugging Face 🤗 Links to repo and weights in 🧵↓ The traditional ASR → LLM → TTS pipeline forces rigid turn-taking. It’s efficient, but it never feels natural. PersonaPlex-7B changes that. This NVIDIA model can listen and speak at the same time. It runs directly on continuous audio tokens with a dual-stream transformer, generating text and audio in parallel instead of passing control between components. That unlocks: → instant back-channel responses → interruptions that feel human → real conversational rhythm Persona control is fully zero-shot! If you’re building low-latency assistants or support agents, this is a big step forward 🔥

Charly Wargnier

564,995 次观看 • 5 个月前

Text-to-speech is moving way too fast. Just a few days ago, I tweeted about PersonaPlex-7B, NVIDIA's new open source TTS ( And today, Qwen just open-sourced Qwen3-TTS 🤯 It’s a revolutionary text-to-speech model built for control. Not just about generating speech, but about shaping how it sounds directly from language. You can guide the pace, the tone, and the expressiveness straight from text, without touching audio graphs or hand-tuning parameters. That’s the real shift! What makes Qwen3-TTS stand out is how practical it already is: → voice cloning from just a few seconds of audio → voice creation without any reference sample → support for 10 languages out of the box → end-to-end latency down to ~97ms → works in both streaming and non-streaming setups The models come in two sizes (0.6B and 1.7B), so you can trade off quality and hardware cost depending on your setup. You can work with curated voices, designed voices, or cloned ones, and it integrates cleanly with vLLM for production use. It also ships as a simple Python package you can pip install. If you’re building real-time voice systems, this removes a lot of friction! 100% free and open source. I put the repo in the 🧵↓

Text-to-speech is moving way too fast. Just a few days ago, I tweeted about PersonaPlex-7B, NVIDIA's new open source TTS ( And today, Qwen just open-sourced Qwen3-TTS 🤯 It’s a revolutionary text-to-speech model built for control. Not just about generating speech, but about shaping how it sounds directly from language. You can guide the pace, the tone, and the expressiveness straight from text, without touching audio graphs or hand-tuning parameters. That’s the real shift! What makes Qwen3-TTS stand out is how practical it already is: → voice cloning from just a few seconds of audio → voice creation without any reference sample → support for 10 languages out of the box → end-to-end latency down to ~97ms → works in both streaming and non-streaming setups The models come in two sizes (0.6B and 1.7B), so you can trade off quality and hardware cost depending on your setup. You can work with curated voices, designed voices, or cloned ones, and it integrates cleanly with vLLM for production use. It also ships as a simple Python package you can pip install. If you’re building real-time voice systems, this removes a lot of friction! 100% free and open source. I put the repo in the 🧵↓

Charly Wargnier

59,144 次观看 • 5 个月前

Mistral just released Voxtral TTS, an open-weight voice model that tries to match ElevenLabs while keeping it free. A 3-billionn param model that fits on a laptop and runs 6X faster than real-time speech. It is split into 3 parts that handle different jobs. - The 3.4B part is the main language brain that understands the text and plans what should be spoken. - The 390M part turns that plan into speech features like rhythm, pronunciation, and tone. - The 300M part converts those speech features into the final compressed audio signal you can play. That split helps because each part only does the job it is best at, which can make the whole system faster and lighter than one giant model doing everything alone. Mistral claims about 90ms to first audio, about 6x faster than real-time speech output, about 3GB RAM after quantization, 9 languages, and voice copying from 5s of audio, even across languages. Its internal tests report 62.8% preference over ElevenLabs Flash on default voices and 69.9% on custom voices, but the bigger point is that companies can keep speech data on their own machines. IMO, enterprise buyers usually care most about control, cost, and latency, So this model should give it great push for voice-based AI apps/products.

Mistral just released Voxtral TTS, an open-weight voice model that tries to match ElevenLabs while keeping it free. A 3-billionn param model that fits on a laptop and runs 6X faster than real-time speech. It is split into 3 parts that handle different jobs. - The 3.4B part is the main language brain that understands the text and plans what should be spoken. - The 390M part turns that plan into speech features like rhythm, pronunciation, and tone. - The 300M part converts those speech features into the final compressed audio signal you can play. That split helps because each part only does the job it is best at, which can make the whole system faster and lighter than one giant model doing everything alone. Mistral claims about 90ms to first audio, about 6x faster than real-time speech output, about 3GB RAM after quantization, 9 languages, and voice copying from 5s of audio, even across languages. Its internal tests report 62.8% preference over ElevenLabs Flash on default voices and 69.9% on custom voices, but the bigger point is that companies can keep speech data on their own machines. IMO, enterprise buyers usually care most about control, cost, and latency, So this model should give it great push for voice-based AI apps/products.

Rohan Paul

15,672 次观看 • 3 个月前

Ok, Kokoro is insane. 🤯 This AI is a groundbreaking TTS model with just 82M parameters. It outperforms larger models and generates minutes of speech in seconds. Oh, and it’s open source! Try it and see for yourself: 👇

Ok, Kokoro is insane. 🤯 This AI is a groundbreaking TTS model with just 82M parameters. It outperforms larger models and generates minutes of speech in seconds. Oh, and it’s open source! Try it and see for yourself: 👇

Min Choi

418,183 次观看 • 1 年前

Kyutai released their Streaming Text to Speech model, ~2B param model, ultra low latency (220ms), CC-BY-4.0 license 🔥 Trained on 2.5 Million Hours of audio, it can serve up to 32 users w/ less than 350ms latency on a SINGLE L40 🤯 Incredible release by kyutai folks, go check out their hugging face page now!

Kyutai released their Streaming Text to Speech model, ~2B param model, ultra low latency (220ms), CC-BY-4.0 license 🔥 Trained on 2.5 Million Hours of audio, it can serve up to 32 users w/ less than 350ms latency on a SINGLE L40 🤯 Incredible release by kyutai folks, go check out their hugging face page now!

Vaibhav (VB) Srivastav

93,512 次观看 • 1 年前

Kokoro is insane. 🤯 This AI is a groundbreaking TTS model with just 82M parameters. It outperforms larger models and generates minutes of speech in seconds. APACHE 2.0 licensed, trained on < 100 hours of audio is now available in ai-gradio pip install --upgrade "ai-gradio[kokoro]" import gradio as gr import ai_gradio gr.load(name='kokoro:kokoro-v0_19', src=ai_gradio.registry,).launch()

Kokoro is insane. 🤯 This AI is a groundbreaking TTS model with just 82M parameters. It outperforms larger models and generates minutes of speech in seconds. APACHE 2.0 licensed, trained on < 100 hours of audio is now available in ai-gradio pip install --upgrade "ai-gradio[kokoro]" import gradio as gr import ai_gradio gr.load(name='kokoro:kokoro-v0_19', src=ai_gradio.registry,).launch()

AK

114,000 次观看 • 1 年前

Pretty WILD - SoTA open source TTS model that beats ElevenLabs/ Sesame - Dia 1.6B - Apache 2.0 licensed! 🔥 > Ultra realistic voice synthesis > Capable of producing non-verbal sounds - coughing, laughing 💥 > Zero shot Voice Cloning > Real-time TTS synthesis > Can run on your MacBook > Trending #2 on Hugging Face Weights on the Hub and code on GitHub! 🤯

Pretty WILD - SoTA open source TTS model that beats ElevenLabs/ Sesame - Dia 1.6B - Apache 2.0 licensed! 🔥 > Ultra realistic voice synthesis > Capable of producing non-verbal sounds - coughing, laughing 💥 > Zero shot Voice Cloning > Real-time TTS synthesis > Can run on your MacBook > Trending #2 on Hugging Face Weights on the Hub and code on GitHub! 🤯

Vaibhav (VB) Srivastav

39,165 次观看 • 1 年前

Pocket TTS goes multilingual! Now you can use our 100M-parameter models to generate speech in six languages, fast enough to run real-time without a GPU. We also improved the quality of the English model while keeping the same size. And all of this is open-source.

Pocket TTS goes multilingual! Now you can use our 100M-parameter models to generate speech in six languages, fast enough to run real-time without a GPU. We also improved the quality of the English model while keeping the same size. And all of this is open-source.

kyutai

29,309 次观看 • 2 个月前

Next gen AI audio separation is here 🤯 AudioSep is a model that can separate audio events, musical instruments, and even enhance speech with natural language queries which makes this a versatile tool for different audio tasks.

Next gen AI audio separation is here 🤯 AudioSep is a model that can separate audio events, musical instruments, and even enhance speech with natural language queries which makes this a versatile tool for different audio tasks.

Dreaming Tulpa 🥓👑

126,029 次观看 • 2 年前

Big News! Meta just released Segment Anything, a new AI model that can "cut out" any object, in any image/video, with a single click. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks.

Big News! Meta just released Segment Anything, a new AI model that can "cut out" any object, in any image/video, with a single click. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks.

Lior Alexander

290,190 次观看 • 3 年前

.Bill Gurley “I don’t think politicians understand how powerful open source is as a competitive weapon. If you have ten companies all doing open source, the minute an idea is shared, it's available to the other ten and they absorb it immediately. I think the odds that the U.S., with two proprietary models, can stay ahead of China, with ten open source ones, is very low.”

.Bill Gurley “I don’t think politicians understand how powerful open source is as a competitive weapon. If you have ten companies all doing open source, the minute an idea is shared, it's available to the other ten and they absorb it immediately. I think the odds that the U.S., with two proprietary models, can stay ahead of China, with ten open source ones, is very low.”

Codie Sanchez

65,669 次观看 • 3 个月前

Calling a model "open source" because you can use it is like calling a car open source because you can drive it. The real question is whether you can open the hood. Azibi Malawi from DataVLT on what openness at the model layer actually requires.

Calling a model "open source" because you can use it is like calling a car open source because you can drive it. The real question is whether you can open the hood. Azibi Malawi from DataVLT on what openness at the model layer actually requires.

Cluster Protocol

26,811 次观看 • 12 天前