Загрузка видео...

Не удалось загрузить видео

На главную

We released Qwen3-Omni-Flash (2025-12-01 version) API Service. Smarter interaction, more human expression: · A/V Interaction: Significant boost in instruction following. Solves "dumbing down" in casual chats with rock-solid stability. · Precise Control: Enhanced System Prompt adherence for specific personas, styles, and lengths. · Multilingual Mastery: Solved language switching instability....

Ali_TongyiLab's profile picture

Tongyi Lab

26,012 subscribers

1,637,135 просмотров • 11 месяцев назад •via X (Twitter)

ОбразованиеНовости и политикаНаука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 27

Фото профиля Tongyi Lab
Tongyi Lab11 месяцев назад

Grab it here： 🎙️ Qwen Chat: click the VoiceChat and VideoChat button (bottom-right): 📝 Blog: 🎧 Demo: 🎧 Demo: ⚡ Realtime API： 📥 Offline API：

Фото профиля Joyce Brown
Joyce Brown11 месяцев назад

Totally resonate with this! 📖 I once struggled with A/V setups, now I'm excited for smoother convos with updates like these!

Фото профиля rohitsankhla3151@gma
rohitsankhla3151@gma10 месяцев назад

Okay

Фото профиля Nolan Spencer
Nolan Spencer10 месяцев назад

Interesting point! 🤔 How does it handle nuances? Any thoughts?

Фото профиля Hyperstack
Hyperstack11 месяцев назад

Big win!

Фото профиля CZ
CZ10 месяцев назад

技术进步确实挺快的，期待看到更多应用

Фото профиля john carter
john carter10 месяцев назад

🤩

Фото профиля Ji'zarr Has Coins
Ji'zarr Has Coins10 месяцев назад

Rock-solid stability and true multilingual mastery? Qwen3-Omni-Flash is shaping up as a game-changer for global human-AI communication.

Фото профиля PrismIon89
PrismIon8910 месяцев назад

ngl AI hype is getting wild... but where's the cash flow? reminds me why we focus on durable assets in Module 3. tech without profit is just noise 💸

Фото профиля vipulbamaniya
vipulbamaniya10 месяцев назад

Awesome

Фото профиля RobbinxApe
RobbinxApe10 месяцев назад

so crazy lol this thing just went to another level

Фото профиля Nelson Wuens | ETHGas ⛽
Nelson Wuens | ETHGas ⛽10 месяцев назад

@aixc_official and @aixcfoundation

Фото профиля Sosaingram
Sosaingram11 месяцев назад

Human speech finally

Фото профиля がーくんです
がーくんです10 месяцев назад

このテーマは素晴らしいですね！自分は✌️毎日少しずつ練習してます。効果はまぁまぁ！もっと良い方法があれば教えてね！

Фото профиля Toromoro
Toromoro10 месяцев назад

precise control - plus! Waiting for test, how does multilingual work?

Фото профиля FranksCannon
FranksCannon10 месяцев назад

so crazy lol this thing just went to another level

Фото профиля fatih
fatih11 месяцев назад

wild

Фото профиля 熊田祥さん
熊田祥さん10 месяцев назад

私も最近よくそう感じる😅

Фото профиля LittleDog04
LittleDog0410 месяцев назад

Good work

Фото профиля ミルクちゃん
ミルクちゃん10 месяцев назад

指示を正確に理解できるって素晴らしいですね！私も体験したい。

Фото профиля やんやん（じゅくとー）@怪談も語る塾の先生
やんやん（じゅくとー）@怪談も語る塾の先生10 месяцев назад

こんばんは🍵 体も心も冷やさずに。

Фото профиля Raian Bin (❖,❖)
Raian Bin (❖,❖)10 месяцев назад

Nice

Фото профиля Zaman
Zaman10 месяцев назад

Cryptocurrency

Фото профиля あーー
あーー10 месяцев назад

確かに両面がありますね⚖️。悪い点も考慮すべきです。

Фото профиля حرية بلدي
حرية بلدي10 месяцев назад

تحیاتنا للجمیع للاطلاع وشکرا

Фото профиля Investman
Investman10 месяцев назад

Big big News

Фото профиля 対ちゃん
対ちゃん10 месяцев назад

共有してくれてありがとう！新しい機能に期待大ですね。皆さんはどう感じますか？

Похожие видео

Flux Multilingual is live. Real-time conversational speech-to-text for voice agents in 10 languages, with monolingual-grade accuracy, turn detection, and code-switching. Deploy once and launch globally. Learn more →
1:13

Flux Multilingual is live. Real-time conversational speech-to-text for voice agents in 10 languages, with monolingual-grade accuracy, turn detection, and code-switching. Deploy once and launch globally. Learn more →

Deepgram

15,930 просмотров • 6 месяцев назад

What if voice AI could understand context, intent, and emotion — not just words? 🎙️ Meet Seed Speech 2.0. A new speech AI stack designed for natural, expressive conversations — with major upgrades in both text-to-speech (TTS) and speech recognition (ASR). Built for developers and creators who want natural conversational speech. What’s inside: - Natural and expressive speech generation - Prompt-controlled emotion and tone - High-accuracy multilingual recognition - Strong contextual reasoning Built for: 🎧 Voice content creation 🤖 AI assistants ☎️ Customer service 🎬 Dubbing & subtitling 📊 Audio-video analysis Voice AI built for expression, accuracy, and understanding. Try it now → Read more about Seed Speech here → #SeedSpeech #VoiceAI #SpeechAI #AI
0:30

What if voice AI could understand context, intent, and emotion — not just words? 🎙️ Meet Seed Speech 2.0. A new speech AI stack designed for natural, expressive conversations — with major upgrades in both text-to-speech (TTS) and speech recognition (ASR). Built for developers and creators who want natural conversational speech. What’s inside: - Natural and expressive speech generation - Prompt-controlled emotion and tone - High-accuracy multilingual recognition - Strong contextual reasoning Built for: 🎧 Voice content creation 🤖 AI assistants ☎️ Customer service 🎬 Dubbing & subtitling 📊 Audio-video analysis Voice AI built for expression, accuracy, and understanding. Try it now → Read more about Seed Speech here → #SeedSpeech #VoiceAI #SpeechAI #AI

BytePlus

19,877 просмотров • 7 месяцев назад

Meet the new Qwen3-TTS (2025-11-27): a major step forward in lifelike voice generation! > 49+ distinctive voices ranging from playful to authoritative, giving creators precise control over personality and style. > Global language coverage with 10 languages and multiple authentic dialects, including Minnan, Wu, Cantonese, Sichuan, Beijing, Nanjing, Tianjin, and Shaanxi. > Human-level delivery with adaptive rhythm, pacing, and intonation that makes speech sound genuinely performed.
0:26

Meet the new Qwen3-TTS (2025-11-27): a major step forward in lifelike voice generation! > 49+ distinctive voices ranging from playful to authoritative, giving creators precise control over personality and style. > Global language coverage with 10 languages and multiple authentic dialects, including Minnan, Wu, Cantonese, Sichuan, Beijing, Nanjing, Tianjin, and Shaanxi. > Human-level delivery with adaptive rhythm, pacing, and intonation that makes speech sound genuinely performed.

steven

45,940 просмотров • 11 месяцев назад

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.
4:10

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.

AK

290,729 просмотров • 3 лет назад

Speech to Speech is now available in 29 languages. In November, we launched Speech to Speech, enabling you to transform your voice into another character with full control over emotions, timing, and delivery. Today we're making Speech to Speech multilingual. Use Speech to Speech to get more control than with prompting alone. Simply say it how you want it, choose a target voice, and generate.
0:37

Speech to Speech is now available in 29 languages. In November, we launched Speech to Speech, enabling you to transform your voice into another character with full control over emotions, timing, and delivery. Today we're making Speech to Speech multilingual. Use Speech to Speech to get more control than with prompting alone. Simply say it how you want it, choose a target voice, and generate.

ElevenLabs

113,343 просмотров • 2 лет назад

Today, we’re excited to introduce Miso One, the most emotive voice model in the world. Miso One is an 8-billion-parameter text-to-speech model for highly expressive speech generation. It emotes like a human and responds faster than a human, with just 110 milliseconds of latency. We’ve open-sourced the model weights, with API access coming soon. Hear how Miso One sounds in the thread below.
0:57

Today, we’re excited to introduce Miso One, the most emotive voice model in the world. Miso One is an 8-billion-parameter text-to-speech model for highly expressive speech generation. It emotes like a human and responds faster than a human, with just 110 milliseconds of latency. We’ve open-sourced the model weights, with API access coming soon. Hear how Miso One sounds in the thread below.

Aoden Teo

5,305,983 просмотров • 5 месяцев назад

Meet the new Qwen3‑TTS lineup: VoiceDesign and VoiceClone • VoiceDesign: controllable speech via text; no presets; strong role‑play performance • VoiceClone: clone from 3s audio; supports 10 languages; lower WER; more natural cadence Follow us for more updates!
3:21

Meet the new Qwen3‑TTS lineup: VoiceDesign and VoiceClone • VoiceDesign: controllable speech via text; no presets; strong role‑play performance • VoiceClone: clone from 3s audio; supports 10 languages; lower WER; more natural cadence Follow us for more updates!

Alibaba Cloud

204,765 просмотров • 10 месяцев назад

HUMAIN Voice is built for how Arabic speakers actually communicate, across dialects, natural code-switching and everyday conversations. Bringing speech-to-text, text-to-speech and agentic voice interaction into one technology family, it enables more natural voice experiences for enterprises, developers and everyday users. Read more: #HUMAIN #LEAP26 #TheEndOfLimits
1:15

HUMAIN Voice is built for how Arabic speakers actually communicate, across dialects, natural code-switching and everyday conversations. Bringing speech-to-text, text-to-speech and agentic voice interaction into one technology family, it enables more natural voice experiences for enterprises, developers and everyday users. Read more: #HUMAIN #LEAP26 #TheEndOfLimits

HUMAIN

21,632 просмотров • 2 месяцев назад

We have officially launched Fun-ASR1.5, a major update to our end-to-end speech recognition model. This release focuses on three core pillars: broader language coverage, language switching, and production-ready text output. Key Features: • Multilingual Support: Supports high-accuracy recognition for 30 languages across Asia, Europe, and the Middle East within a single model. • Language Switching: Handles mixed-language speech (Code-Switching) natively, automatically detecting and transcribing language shifts without the need for manual tagging. • Professional Text Output: Delivers "ready-to-use" text with smart punctuation and automatic formatting for dates, numbers, and currencies. Fun-ASR1.5 bridges the gap between raw audio and professional documentation, providing a reliable engine for global communication.
1:35

We have officially launched Fun-ASR1.5, a major update to our end-to-end speech recognition model. This release focuses on three core pillars: broader language coverage, language switching, and production-ready text output. Key Features: • Multilingual Support: Supports high-accuracy recognition for 30 languages across Asia, Europe, and the Middle East within a single model. • Language Switching: Handles mixed-language speech (Code-Switching) natively, automatically detecting and transcribing language shifts without the need for manual tagging. • Professional Text Output: Delivers "ready-to-use" text with smart punctuation and automatic formatting for dates, numbers, and currencies. Fun-ASR1.5 bridges the gap between raw audio and professional documentation, providing a reliable engine for global communication.

Tongyi Lab

3,917,307 просмотров • 6 месяцев назад

Meet MiniMax Speech 2.6 — Ultra-Fast, Ultra-Human, Ultra-Smart ⚡Ultra-Fast: <250 ms latency for real-time conversational interactions 💬Smart Text Normalization for URLs, emails, dates, numbers & more 🎙️Full Voice Clone + Fluent LoRA: natural, expressive, effortless 🌍40+ languages with inline code switching Fluent anytime. In every language. 👉 API available:
1:02

Meet MiniMax Speech 2.6 — Ultra-Fast, Ultra-Human, Ultra-Smart ⚡Ultra-Fast: <250 ms latency for real-time conversational interactions 💬Smart Text Normalization for URLs, emails, dates, numbers & more 🎙️Full Voice Clone + Fluent LoRA: natural, expressive, effortless 🌍40+ languages with inline code switching Fluent anytime. In every language. 👉 API available:

Hailuo AI (MiniMax)

553,377 просмотров • 1 год назад

Play AI by vibx (YC W23) is a new platform for creating human-like voice agents. It makes speech generation contextual and handles turn-taking, interruption, voice energy, and emotion modulation for natural, fluid, human conversations in real-time.
2:02

Play AI by vibx (YC W23) is a new platform for creating human-like voice agents. It makes speech generation contextual and handles turn-taking, interruption, voice energy, and emotion modulation for natural, fluid, human conversations in real-time.

Y Combinator

21,285 просмотров • 2 лет назад

Introducing — a new conversational Voice AI platform and API for creating human-like voice agents and voice interfaces. Powered by our new Large Dialogue Model (LDM), Play LDM 1.0 makes the speech generation contextual; handles turn-taking, interruption, voice energy and emotion modulation for a natural, fluid, human conversations in real-time.
2:02

Introducing — a new conversational Voice AI platform and API for creating human-like voice agents and voice interfaces. Powered by our new Large Dialogue Model (LDM), Play LDM 1.0 makes the speech generation contextual; handles turn-taking, interruption, voice energy and emotion modulation for a natural, fluid, human conversations in real-time.

PlayAI

105,692 просмотров • 2 лет назад

Today, we're introducing Bulbul V4 Flash, our latest text to speech model built for natural, expressive speech across Indian languages. Flash brings frontier-quality speech into a smaller, more efficient model, preserving the subtle details that make voices sound natural while making speech generation significantly faster. See details in our blog:
1:18

Today, we're introducing Bulbul V4 Flash, our latest text to speech model built for natural, expressive speech across Indian languages. Flash brings frontier-quality speech into a smaller, more efficient model, preserving the subtle details that make voices sound natural while making speech generation significantly faster. See details in our blog:

Sarvam

161,705 просмотров • 28 дней назад

Mini-Omni 2 understands image, audio and text inputs all via end-to-end voice conversations with users 🔥 > Understands and processes images, speech, and text > Generates real-time speech responses > Supports interruptions during speech Technical Overview: > Concatenates image, audio, and text features for input. > Uses text-guided delayed parallel output for real-time speech > Involves encoder adaptation, modal alignment, and multimodal fine-tuning Best part: MIT licensed ⚡
0:54

Mini-Omni 2 understands image, audio and text inputs all via end-to-end voice conversations with users 🔥 > Understands and processes images, speech, and text > Generates real-time speech responses > Supports interruptions during speech Technical Overview: > Concatenates image, audio, and text features for input. > Uses text-guided delayed parallel output for real-time speech > Involves encoder adaptation, modal alignment, and multimodal fine-tuning Best part: MIT licensed ⚡

Vaibhav (VB) Srivastav

45,453 просмотров • 2 лет назад

Introducing Saaras V4, our most capable speech to text model yet. It delivers strong performance across both English and Indian languages, with better accuracy over a much wider range of speech. Saaras V4 was built with specific attention to noise, accents, dialects and mixed-language speech. Read more in the blog:
1:13

Introducing Saaras V4, our most capable speech to text model yet. It delivers strong performance across both English and Indian languages, with better accuracy over a much wider range of speech. Saaras V4 was built with specific attention to noise, accents, dialects and mixed-language speech. Read more in the blog:

Sarvam

219,621 просмотров • 1 месяц назад

New tiny VLM: LFM2.5-VL-450M &gt; Supports bounding box prediction, object detection, and function calling &gt; Improved multilingual capabilities across 9 languages &gt; Enhanced instruction following for vision and text tasks
1:14

New tiny VLM: LFM2.5-VL-450M &gt; Supports bounding box prediction, object detection, and function calling &gt; Improved multilingual capabilities across 9 languages &gt; Enhanced instruction following for vision and text tasks

Maxime Labonne

32,242 просмотров • 7 месяцев назад

Grok's Text to Speech API is now available in LiveKit Inference. Natural, expressive voices with low-latency streaming. Multilingual in 20+ languages. Telephony and production-ready out of the box. One API key. No extra setup. →
0:42

Grok's Text to Speech API is now available in LiveKit Inference. Natural, expressive voices with low-latency streaming. Multilingual in 20+ languages. Telephony and production-ready out of the box. One API key. No extra setup. →

LiveKit

159,685 просмотров • 7 месяцев назад

Listen up 🔊 We’ve made some updates to our Gemini Audio models and capabilities: — Gemini’s live speech-to-speech translation capability is rolling out in a beta experience to the Google Translate app, bringing you real-time audio translation that captures the nuance of human speech — Gemini 2.5 Flash and 2.5 Pro Text-to-Speech preview models bring improved adherence to style prompts, precision pacing with context-aware speed adjustments, and character voice consistency for multi-speaker scenarios — Gemini 2.5 Flash Native Audio is now updated, with improvements to handle complex workflows, navigate user instructions, and hold natural conversations
2:50

Listen up 🔊 We’ve made some updates to our Gemini Audio models and capabilities: — Gemini’s live speech-to-speech translation capability is rolling out in a beta experience to the Google Translate app, bringing you real-time audio translation that captures the nuance of human speech — Gemini 2.5 Flash and 2.5 Pro Text-to-Speech preview models bring improved adherence to style prompts, precision pacing with context-aware speed adjustments, and character voice consistency for multi-speaker scenarios — Gemini 2.5 Flash Native Audio is now updated, with improvements to handle complex workflows, navigate user instructions, and hold natural conversations

Google AI

1,095,837 просмотров • 11 месяцев назад

Day 1 of 3 MLX Releases: Introducing MLX-Audio 🚀🔥 A text-to-speech (TTS) and Speech-to-Speech (STS) library built on Apple's MLX framework, providing efficient speech synthesis on Apple Silicon. Features ⚡️Fast inference on Apple Silicon (M series chips) 🤖Multiple language support 🗣️Voice customization options 🚀Quantization support for optimized performance Supported models: 🪶Kokoro - A multilingual TTS model with 82M params that supports various languages and voice styles. With more models coming soon. Get started: > pip install mlx-audio Please leave us a star and send a PR :)
4:07

Day 1 of 3 MLX Releases: Introducing MLX-Audio 🚀🔥 A text-to-speech (TTS) and Speech-to-Speech (STS) library built on Apple's MLX framework, providing efficient speech synthesis on Apple Silicon. Features ⚡️Fast inference on Apple Silicon (M series chips) 🤖Multiple language support 🗣️Voice customization options 🚀Quantization support for optimized performance Supported models: 🪶Kokoro - A multilingual TTS model with 82M params that supports various languages and voice styles. With more models coming soon. Get started: > pip install mlx-audio Please leave us a star and send a PR :)

Prince Canuma

123,553 просмотров • 1 год назад

Introducing Gemini 3.5 Transcribe, our most precise speech-to-text model yet, designed for intelligent voice interactions! 💬 Available for both real-time speech and audio file transcription, with support for: 💡 smart transcription: rewrite and format text based on intent. 🌍 support for 85+ languages, regional accents and diverse dialects. 📚 custom vocabulary: recognize jargon and specific spellings. 💯 high precision in alphanumerics. 🫂 multi-speaker identification.
4:49

Introducing Gemini 3.5 Transcribe, our most precise speech-to-text model yet, designed for intelligent voice interactions! 💬 Available for both real-time speech and audio file transcription, with support for: 💡 smart transcription: rewrite and format text based on intent. 🌍 support for 85+ languages, regional accents and diverse dialects. 📚 custom vocabulary: recognize jargon and specific spellings. 💯 high precision in alphanumerics. 🫂 multi-speaker identification.

Thor 雷神 ⚡️

23,123 просмотров • 2 месяцев назад