---
title: "No Text at All: Twitter Founder Trains AI to Understand Human Speech by Ear"
description: "Jack Dorsey's Block has created the Dessa voice AI model trained entirely on audio without text. The system engages in real-time conversation while capturing intonations and emotions."
date: 2026-10-26T12:20:31.634Z
lang: en
url: https://xab.info/en/posts/twitter-founder-trains-ai-speech-by-ear
tags: [artificial-intelligence, jack-dorsey, audio-ai, block, machine-learning]
author: "Alexander Buran"
publisher: "XAB.info"
---

# No Text at All: Twitter Founder Trains AI to Understand Human Speech by Ear

![Twitter founder Jack Dorsey in a black beanie against the background of a digital sound wave with the inscription dessa](https://xab.info/media/2026/10/26/osnovatel-twitter-nauchil-ii-ponimat-yazyk-na-sluh/osnovatel-twitter-nauchil-ii-ponimat-yazyk-na-sluh-1.webp)

## Key Points

- The Dessa model was trained exclusively on audio recordings without using text
- The system operates on direct audio signals without intermediate text recognition
- Full-duplex mode is supported with simultaneous listening and speaking capabilities
- The artificial intelligence can run locally on MacBook Pro laptops with M4 Max chips

Block, the company founded by former Twitter executive Jack Dorsey, has introduced a new voice AI model called Dessa. The main feature of the development is the complete rejection of text data during training: the system perceives and processes exclusively audio recordings of human conversations. According to the creators, this approach allows the model to capture the smallest nuances of spoken speech, including intonation, pauses, and emotional coloring, ensuring the most natural interaction possible.

### Architecture Without a Text Layer

Traditional voice assistants and modern conversational AI models typically use a multi-stage architecture. First, the heard sound is converted into text using speech recognition modules, then the text request is processed by a language model, and finally, the response is synthesized back into a voice format. This approach inevitably leads to the loss of emotional shades, intonations, and the natural rhythm of the conversation.

Unlike classical systems, the Dessa model operates on direct audio signals without an intermediate text stage. This allows the artificial intelligence not only to hear the interlocutor's words but also to simultaneously analyze their emotional state. The system is capable of instantly reacting to sarcasm, laughter, voice fluctuations, or changes in tone right during the conversation, making communication virtually indistinguishable from a dialogue with a living person.

### Full-Duplex Capabilities

The new development supports full-duplex communication, which means the ability to listen and speak simultaneously. Thanks to this, the artificial intelligence can pause in time, react to the interlocutor's remarks, or even interrupt them depending on the context of the conversation. During internal testing, the system demonstrated the ability to independently supplement forgotten movie titles or change its speech manner upon the user's request.

In addition to high adaptability, the model stands out for its hardware efficiency. To train it, engineers used only a single high-performance server with graphics processors, and the finished version is capable of operating locally on MacBook Pro laptops equipped with Apple M4 Max chips and higher. The audio processing speed exceeds real-time, completely eliminating delays in dialogue.

### Release Prospects and Availability

At present, the Dessa voice model has not been released into open access for the general audience. The developers provide limited testing access exclusively to artificial intelligence safety specialists to conduct independent audits and checks. Other interested users can sign up for the waiting list on the company's official resources while the team completes preparations for a full-scale commercial launch.

## Fact-Check & Sources

- [Ни одного текста: основатель Twitter научил ИИ понимать человеческий язык на слух](https://www.rbc.ua/ukr/news/zhodnogo-tekstu-zasnovnik-twitter-navchiv-1793010930.html) - The article is fully based on the RBC-Ukraine publication regarding the announcement of the Dessa voice model by Jack Dorsey on X.

## FAQ

### What is the main difference between the Dessa model and classic voice assistants?

Dessa is trained exclusively on human speech audio recordings and operates on direct audio signals, bypassing the speech-to-text conversion stage.

### What technical requirements are needed to run the model?

The finished model can run entirely locally on MacBook Pro laptops equipped with an Apple M4 Max processor or more powerful chips.

### Is the new development available to the general public?

Currently, access is restricted to AI safety specialists, while other interested users can join a waiting list.