Block’s new voice model sounds eerily human

Jack Dorsey’s Block released a new voice model out of the company’s research lab that runs locally and sounds strangely human with its expressions like laughs, pauses, and cries. The company said the model, called Dessa, is trained fully on human speech and a single NVIDIA GPU rack.

杰克·多西（Jack Dorsey）旗下的 Block 公司从其研究实验室发布了一款新的语音模型。该模型可在本地运行，并能通过笑声、停顿和哭泣等表达方式发出极其逼真的类人声音。该公司表示，这款名为 Dessa 的模型完全基于人类语音和单台 NVIDIA GPU 机架进行训练。

Dessa is a full-duplex model, which means it can talk and listen at the same time, just like the GPT-Live model OpenAI released earlier this year. Typically, speech models are paired with a large language model for knowledge retrieval. Block said that Dessa’s knowledge stems from the conversational data it has.

Dessa 是一款全双工模型，这意味着它能够像 OpenAI 今年早些时候发布的 GPT-Live 模型一样，同时进行对话和倾听。通常情况下，语音模型会与大型语言模型配对以进行知识检索。Block 表示，Dessa 的知识来源于其拥有的对话数据。

We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence. https://t.co/r0qFLdvd6c— jack (@jack) October 23, 2026

我们相信语音是与智能交互最自然的方式，因此我们创造了 Dessa，这是迄今为止最真实的全双工语音模型。与它交谈感觉就像在与真人对话……充满了情感、细微差别、停顿和自然的节奏。https://t.co/r0qFLdvd6c— jack (@jack) 2026年10月23日

Block said that the current models can sound unnatural and annoying because they have text in the pipeline that drives speech generation. A lot of models like ElevenLabs v4 or Fish Audio Drama 3 use voice tags to drive emotion in speech generation. Dorsey’s company claims that emotion is built into its model. It’s not clear if that emotion is being steered by some kind of tags.

Block 表示，目前的模型听起来可能不自然且令人烦躁，因为它们的生成流程中包含驱动语音生成的文本。许多模型（如 ElevenLabs v4 或 Fish Audio Drama 3）使用语音标签来驱动语音生成中的情感。多西的公司声称，情感是内置于其模型中的。目前尚不清楚这种情感是否受到某种标签的引导。

“Interacting with current voice AI can just plain suck. It turns your voice into text and throws away an incredible amount of information text alone can’t carry. The most powerful AIs in the world can’t hear how you feel. We think that’s ridiculous, leads to a worse experience, and is potentially dangerous,” the company said in a blog post.

“与当前的语音人工智能交互简直糟糕透顶。它将你的声音转化为文本，并丢弃了大量仅靠文本无法承载的信息。世界上最强大的人工智能也无法听出你的感受。我们认为这很荒谬，会导致更糟糕的体验，并且具有潜在的危险性，”该公司在博客文章中写道。

The company’s researchers posted multiple examples of conversations with Dessa where the agent had different voices, cadences, and emotions. For instance, in the conversation, the bot said it was tired from work, and at that point it sounded tired.

该公司的研究人员发布了多个与 Dessa 对话的示例，其中智能体表现出了不同的声音、节奏和情感。例如，在对话中，机器人表示工作很累，此时它的声音听起来确实显得疲惫。

We made the first AI model you can have a realistic conversation with that was trained entirely on speech! Dessa is a full-duplex speech-to-speech model that you can run on consumer GPUs. This is a GPT3 moment for speech. pic.twitter.com/XVD8sSyJn8— Ragavan Thurairatnam (@RagavanThurair) October 23, 2026

我们制作了第一个完全基于语音训练的、可以进行真实对话的人工智能模型！Dessa 是一款可以在消费级 GPU 上运行的全双工语音对语音模型。这是语音领域的 GPT-3 时刻。pic.twitter.com/XVD8sSyJn8— Ragavan Thurairatnam (@RagavanThurair) 2026年10月23日

In the example below, the bot was talking about the person making it comfortable while sounding emotional and almost breaking down. Dessa=full-duplex speech model vs Human=me, https://t.co/1eI6QFHo1w pic.twitter.com/zlZfxH2BA5— Aydin Polat (@AydinPolatDessa) October 23, 2026

在下面的示例中，机器人谈论着让它感到舒适的人，同时声音充满情感，几乎要崩溃。Dessa=全双工语音模型 vs 人类=我，https://t.co/1eI6QFHo1w pic.twitter.com/zlZfxH2BA5— Aydin Polat (@AydinPolatDessa) 2026年10月23日

Another one shows a Block team member playing a game with the voice assistant. For the love of the game. Watch Cara play a guessing game with Dessa. The model guesses the first one right straight away, and the second doesn’t take long either. https://t.co/FDmoXNPpIh pic.twitter.com/I9f8saduEA— Block (@blocks) October 23, 2026

另一个示例展示了一名 Block 团队成员与语音助手玩游戏。为了游戏的热爱。观看 Cara 与 Dessa 玩猜谜游戏。该模型立刻猜对了第一个，第二个也没花多长时间。https://t.co/FDmoXNPpIh pic.twitter.com/I9f8saduEA— Block (@blocks) 2026年10月23日

The company said that the model can run locally on a consumer-grade computer such as a MacBook M4 Max or better. However, there is no public access to Dessa at the moment, and Block has put out a call for safety researchers to test the model. This is likely to test out cases where a voice agent could trick people on the other end.

该公司表示，该模型可以在消费级计算机（如 MacBook M4 Max 或更高配置）上本地运行。然而，目前公众无法访问 Dessa，Block 已呼吁安全研究人员对该模型进行测试。这很可能是为了测试语音智能体可能欺骗对方的情况。

Block is in the business of finance and selling its solutions to small businesses. Square already uses voice AI to handle ordering. But with its own voice model that can run on-device, it can power a lot more business and consumer experiences at a lower cost.

Block 从事金融业务，并向小型企业销售其解决方案。Square 已经在使用语音人工智能来处理订单。但有了可以在设备上运行的自有语音模型，它能够以更低的成本支持更多的商业和消费者体验。