Unverified50% confidenceReleaseExact time
Tencent open-sourced a voice generation and editing model that can modify recorded content using natural language while preserving the melody and original timbre; the base model has 1.5 billion parameters, and a Flash distilled version with four-step inference is also provided, with code and weights publicly released.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/10/2026
First Seen
Valid until: 1/8/2027
Sources
Related Entities
Related Claims
UnverifiedThe plugin includes Tongyi Qianwen's text-to-speech node, voice design node, and voice cloning node, with the voice cloning node capable of automatically downloading models.74% similarUnverifiedByteDance's Seed Audio is its in-house multimodal audio generation large model, based on diffusion models and large-scale audio pre-training technology.74% similarUnverifiedByteDance Doubao released the Seed Audio 1.0 audio generation model.73% similarVerifiedWhisper is an automatic speech recognition model open-sourced by OpenAI in 2022, built on a Transformer architecture and trained on 680,000 hours of multilingual audio data.73% similarUnverifiedWhisper audio is first converted into an 80-dimensional mel spectrogram, downsampled by convolutional layers, fed into the encoder to generate a feature sequence, and then decoded by the decoder to output text token by token in an autoregressive manner73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/993511API
curl https://kongchang.com/api/v1/knowledge/claims/993511MCP
get_claim(id=993511)