Average Ratings 0 Ratings
Average Ratings 0 Ratings
Description
CosyVoice is a sophisticated voice cloning and speech synthesis model developed by Qwen Cloud, part of the CosyVoice series, which is specifically aimed at enhancing professional applications in text-to-speech with notable improvements in audio quality, naturalness, expressiveness, and cloning accuracy. This model can generate a custom voice that closely resembles the reference audio after a brief recording, requiring just 10–20 seconds of clear speech to achieve optimal results, although a minimum of five seconds of uninterrupted dialogue is essential. It is equipped for real-time streaming text-to-speech synthesis, which enables applications to process text and deliver audio with minimal initial latency. Supporting multiple languages including Chinese, English, French, German, Japanese, Korean, and Russian, the model offers language hints during the enrollment process to facilitate better voice identification. The source recordings accepted by the model can be in WAV, MP3, or M4A formats and should consist of clear speech devoid of any background music, noise, or other speakers to ensure the best possible output. Overall, CosyVoice stands out as a powerful tool for creating personalized voice experiences in various linguistic contexts.
Description
Qwen3-Omni is a comprehensive multilingual omni-modal foundation model designed to handle text, images, audio, and video, providing real-time streaming responses in both textual and natural spoken formats. Utilizing a unique Thinker-Talker architecture along with a Mixture-of-Experts (MoE) framework, it employs early text-centric pretraining and mixed multimodal training, ensuring high-quality performance across all formats without compromising on text or image fidelity. This model is capable of supporting 119 different text languages, 19 languages for speech input, and 10 languages for speech output. Demonstrating exceptional capabilities, it achieves state-of-the-art performance across 36 benchmarks related to audio and audio-visual tasks, securing open-source SOTA on 32 benchmarks and overall SOTA on 22, thereby rivaling or equaling prominent closed-source models like Gemini-2.5 Pro and GPT-4o. To enhance efficiency and reduce latency in audio and video streaming, the Talker component leverages a multi-codebook strategy to predict discrete speech codecs, effectively replacing more cumbersome diffusion methods. Additionally, this innovative model stands out for its versatility and adaptability across a wide array of applications.
API Access
Has API
API Access
Has API
Integrations
ConvNetJS
GPT-4o
Gemini 2.5 Pro
Gemini 2.5 Pro Deep Think
Gemini 3 Deep Think
Hermes Agent
OpenClaw
Qwen
Qwen Studio
QwenCloud
Integrations
ConvNetJS
GPT-4o
Gemini 2.5 Pro
Gemini 2.5 Pro Deep Think
Gemini 3 Deep Think
Hermes Agent
OpenClaw
Qwen
Qwen Studio
QwenCloud
Pricing Details
$0.26 per 10,000 characters
Free Trial
Free Version
Pricing Details
No price information available.
Free Trial
Free Version
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Deployment
Web-Based
On-Premises
iPhone App
iPad App
Android App
Windows
Mac
Linux
Chromebook
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Customer Support
Business Hours
Live Rep (24/7)
Online Support
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Types of Training
Training Docs
Webinars
Live Training (Online)
In Person
Vendor Details
Company Name
Alibaba
Founded
1999
Country
China
Website
www.qwencloud.com/models/cosyvoice-v3-plus
Vendor Details
Company Name
Alibaba
Founded
1999
Country
China
Website
qwen.ai/blog