Google has launched the Pixel 11 series with a major accessibility-focused AI feature that allows users to translate American Sign Language (ASL) into English text directly on their phones. The capability is powered by Google’s new sign-language-to-text (SL2T) model and is integrated into Gboard and Live Transcribe, giving Deaf and hard-of-hearing users a new way to communicate, search the web, write messages, and interact with Gemini without typing.
The feature represents one of Google’s most significant attempts to bring sign-language AI from research into a mainstream consumer product. Rather than treating sign language as a conventional hand-gesture recognition problem, Google’s system is designed to understand broader movements involved in signing, including hand, body and facial cues. The Pixel 11 launch therefore combines Google’s latest smartphone hardware with AI capabilities aimed at making everyday digital communication more accessible.
Pixel 11 Brings Sign Language AI to Gboard
The new SL2T model enables users to sign instead of type when entering text on the Pixel 11.
The feature can be used for tasks such as:
- Searching the web.
- Drafting messages.
- Writing documents.
- Asking Gemini questions.
- Entering text into compatible applications.
- Responding during conversations through Live Transcribe.
In Gboard, the camera can capture a user’s signing and convert it into English text that can then be inserted wherever normal keyboard input would be used. In Live Transcribe, the technology is designed to allow users to sign responses during conversations rather than having to type them manually.
Pixel 11 Accessibility Feature at a Glance
| Feature | Details |
|---|---|
| AI Model | Sign-language-to-text (SL2T) |
| Launch Device | Pixel 11 |
| Initial Language | American Sign Language |
| Output | English text |
| Apps | Gboard and Live Transcribe |
| Additional Integration | Gemini |
How Google’s Sign-to-Text Technology Works
Google’s approach goes beyond simply recognizing individual hand gestures.
The SL2T system is designed to understand the spatial and temporal information involved in signing. This includes movements involving the:
- Hands.
- Arms.
- Face.
- Torso.
- Body position.
This is important because meaning in sign languages can depend on movement, facial expressions and spatial relationships rather than just the shape of a person’s hands at a particular moment.
Google’s technology is also designed to support one-handed and left-handed signing, making the system more flexible for different users.
Privacy Is a Major Part of the System
Google is also emphasizing privacy in the way the technology processes sign language.
According to details about the system, the phone can use MediaPipe Holistic to identify body landmarks on the device. Instead of sending the user’s raw camera footage for processing, the system converts relevant visual information into geometric landmark data.
The approach is intended to reduce the amount of sensitive visual information that leaves the device, while still allowing Google’s AI systems to interpret the signing.
This could be particularly important for an accessibility feature that requires users to point a smartphone camera toward themselves during private conversations.
Initial Support Is Limited to ASL
At launch, the sign-to-text feature supports American Sign Language to English.
Google’s underlying SL2T research covers a much broader range of sign languages. Reports indicate that the model was trained using more than 100,000 hours of data covering more than 50 sign languages, although ASL-to-English is the initial consumer implementation. Additional languages and devices are expected to follow.
Current Availability
| Area | Launch Status |
|---|---|
| American Sign Language | Supported |
| English Text Output | Supported |
| Gboard | Supported |
| Live Transcribe | Supported |
| Gemini Interaction | Supported |
| Other Sign Languages | More languages planned |
| Other Android Devices | More devices planned |
Gemini Adds Another Layer of Accessibility
The integration with Gemini could make the feature more useful than a basic transcription tool.
Once signing has been converted into text, users can use that text to interact with Google’s AI assistant. This could include asking questions, searching for information, drafting content or initiating supported tasks.
The combination effectively turns sign-language input into another interface for interacting with Google’s broader AI ecosystem.
For example, a user could sign a question to the phone, have the system convert it into text, and then use that input with Gemini rather than manually typing the request.
Pixel 11 Combines AI With New Hardware
The sign-language feature is part of a broader AI-focused Pixel 11 launch.
Google’s new smartphone family includes the Pixel 11, Pixel 11 Pro, Pixel 11 Pro XL and Pixel 11 Pro Fold, with the lineup powered by Google’s new Tensor G6 processor. The company is also introducing additional AI-powered capabilities across photography, translation and device interaction.
The Pixel 11 Pro Fold receives particular attention for accessibility because its larger foldable display can provide additional screen space while the camera captures sign language.
Google is also using AI for real-time media translation, including automatic dubbing of supported video and audio content, extending the company’s broader push toward multimodal translation.
A Significant Step for Assistive Technology
The introduction of sign-to-text directly into a consumer smartphone could have broader implications for assistive technology.
Traditional communication between sign-language users and people who do not understand sign language can require typing, interpreters or specialized tools. A smartphone that can translate signing into text could provide an additional option for everyday interactions.
Potential use cases include:
- Conversations between Deaf and hearing people.
- Searching and messaging without typing.
- Communication in public settings.
- Interacting with AI assistants.
- Accessibility-focused education and workplace applications.
However, the technology should not be viewed as a replacement for professional interpreters in situations where accuracy and nuance are critical. Sign languages contain linguistic and cultural characteristics that can be difficult for automated systems to capture perfectly.
Google’s Longer-Term Sign Language Ambitions
The Pixel 11 launch appears to be an important step in Google’s broader effort to commercialize sign-language AI.
Google previously introduced SignGemma, an AI model designed to translate sign language into spoken or written language. The company described it as an open model intended to expand possibilities for communication between people who use sign language and those who do not.
Moving sign-language technology from research models into Gboard and Live Transcribe represents a shift toward making the capability part of everyday smartphone interactions.
The company’s decision to start with ASL also gives Google an opportunity to gather real-world feedback before expanding the feature to additional languages and devices.
Looking Ahead
Google’s Pixel 11 sign-language transcription feature demonstrates how generative and multimodal AI can move beyond conventional productivity applications and address practical accessibility challenges. By allowing users to sign into Gboard and Live Transcribe, Google is effectively introducing sign language as another form of smartphone input, alongside typing, speech and touch. The initial ASL-to-English implementation is limited in scope, but its integration with Gemini gives the technology broader potential across communication and digital services.
Looking ahead, the biggest test will be how accurately and naturally the system performs in real-world environments and how quickly Google expands support to additional sign languages and Android devices. If the technology proves reliable, sign-to-text could become an important accessibility layer across smartphones, AI assistants and other consumer technologies, potentially making digital communication more inclusive for millions of Deaf and hard-of-hearing users.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
