文字起こし翻訳 is an iPhone app that transcribes your voice in real time, translates between Japanese and English, and reads the result aloud — all from a single screen. Tap the record button and speak. Your words appear as text as you talk. Tap Translate to convert the text instantly using Apple's on-device Translation API. Tap Read Aloud to hear the result in a native voice. Switch between Japanese and English with one tap in the navigation bar.
Tap the record button and start talking. SFSpeechRecognizer converts your voice to text in real time, updating the display as you speak so you can follow along. The text is automatically formatted with line breaks at punctuation marks, keeping long passages readable.
Translation is powered by Apple's standard Translation API, available on iOS 18 and later. Tap the Translate button to convert the transcribed text instantly. Switch the translation direction anytime by selecting a language from the flag menu in the navigation bar.
Tap the Read Aloud button and AVSpeechSynthesizer plays the translated text in a native-sounding voice for the target language. The app sets the audio session to playback mode so the voice is heard even when the device is on silent. Tap again to stop instantly.
The flag menu in the navigation bar lets you choose between Japanese (🇯🇵) and English (🇺🇸). Switching updates the speech recognition locale, translation direction, and text-to-speech language code all at once. The menu is disabled during recording to prevent accidental changes.
Tap the record button to start. The microphone captures your voice and SFSpeechRecognizer streams partial results to the screen in real time. Japanese punctuation (。！？、) and English punctuation (. ! ?) each trigger automatic line breaks, so the transcribed text stays easy to read no matter how long you speak. The record button changes to a square stop icon while recording — tap it again to stop.
The app uses the Translation API introduced in iOS 18, which runs on-device for supported language pairs after the initial model download. Tap Translate and the transcribed text is sent to the session, with the result appearing in the lower text area. The Translate button is automatically disabled when the transcription is empty, recording is in progress, or the translation session is not yet ready.
Once a translation is available, tap Read Aloud. AVSpeechSynthesizer reads the text aloud using a voice matched to the target language. The audio session is set to the .playback category so sound plays even when the device is silenced. If the synthesizer is already speaking, tapping Read Aloud stops it immediately before starting the new utterance.
The flag icon in the top-right corner of the navigation bar opens a language menu with Japanese (🇯🇵) and English (🇺🇸). Selecting a language updates the SFSpeechRecognizer locale, the TranslationSession source and target language codes, and the AVSpeechSynthesizer language — everything in one tap. The menu is locked during recording to avoid disrupting an active session.
Tap the flag menu in the navigation bar to select the language you'll be speaking — Japanese or English. Then tap the record button and speak. Your words are transcribed in real time as you talk.
When you're done speaking, tap the Translate button. The transcribed text is translated instantly using Apple's Translation API and displayed in the lower area of the screen.
Tap Read Aloud to hear the translated text played back in a native voice. Even in silent mode the audio plays through clearly, so you can turn the screen toward someone and let them listen directly.
"I use this while traveling abroad to transcribe English and get a Japanese translation. Two button taps after recording gives me the translation and the read-aloud — no fumbling around. Apple's translation quality is solid and I trust it for everyday conversations."
"I record English meetings and use the transcription as a starting point for notes. The automatic line breaks at punctuation make the output much cleaner than other tools I've tried. Simple and reliable."
"I speak Japanese into it, get the English translation, and use the read-aloud to check my pronunciation. The interface is minimal so there's nothing confusing to navigate, and the sound plays even on silent which is a must for me."
Find answers to common questions below.