Speech Recognition & Synthesis: Voice Input and Output

Speech Recognition & Synthesis is a system-level utility app by Google LLC that supplies both speech-to-text and text-to-speech services to other applications on your Android device. Rather than being a standalone productivity tool, it acts as a background engine that other apps call upon whenever they need to convert spoken words into text or read on-screen content aloud. Installing or updating it ensures that voice-dependent features across your device have a consistent, capable recognizer and synthesizer to rely on.

The app serves a wide range of use cases. Navigation apps can understand spoken place names, note-taking apps can transcribe recordings locally on the device, and the Phone app can generate live captions of incoming callers. Accessibility tools such as Voice Access let you operate the phone entirely by voice, while dictation keyboards and language-learning apps use the same speech-to-text engine to recognize what you say. On the text-to-speech side, e-book readers can narrate pages, translation apps can pronounce words aloud, and screen readers like TalkBack provide spoken feedback throughout the interface.

Setting it up is straightforward but requires a small amount of manual configuration. To use the speech-to-text portion, you select the app as your preferred assist or voice input engine in the system settings. For text-to-speech, you choose it as the default output engine in the language and input settings. On many Android devices the service is already present, so installing this version mainly serves as an update that brings the latest improvements.

Key features

• Provides speech-to-text conversion that other apps use for voice search, dictation, and live transcription
• Offers text-to-speech output so apps can read screen content, books, and translations aloud
• Enables real-time call captions through the Phone app Call Screen feature
• Transcribes recordings on the device for supported recorder apps
• Works with accessibility tools like Voice Access and TalkBack for hands-free control
• Uses only the microphone permission to capture audio for recognition
• Has no ads and no in-app purchases, according to the developer
• Runs on Android 8.0 and newer with a 15 MB footprint

How to install

1. Tap the download button on this page and wait for the 15 MB APK file to finish downloading.
2. When Android prompts you, allow your browser or file manager to install apps by enabling the 'Install unknown apps' setting for that app.
3. Open the downloaded APK file from your notifications or file manager.
4. Tap Install and, if asked, confirm the installation.
5. Open the app once installed; you can then set it as your default voice and text-to-speech engine in system settings.

Latest changes

The app now goes by the name "Speech Recognition and Synthesis from Google". In addition to converting on-screen text into speech, it can also turn your spoken words into text for use by other apps. Further details are available in the "About this app" section.

It is for Android users who rely on voice input, dictation, screen readers, or any app that converts speech to text or text to speech, including people with accessibility needs, language learners, and frequent users of navigation, recording, or translation apps.

After installing, check Settings > Apps & notifications > Default apps and Settings > Languages & Input > Text-to-Speech output to assign the app as your default engine. If a voice feature stops responding after an update, restarting the device usually re-registers the services so they work again.

Speech Recognition & Synthesis is an essential system component rather than a flashy standalone app, quietly powering voice features across many tools you already use.