I’m building an iOS app in Swift that should:
- Record user voice 🎙
- Transcribe it to text using SFSpeechRecognizer
- Auto-detect the input language
- Translate it to another language (selected by the user) using a free API (like MyMemory / LibreTranslate)
- Speak back the translated text using AVSpeechSynthesizer
- Show a live animated waveform during recording
- Handle microphone + speech permissions gracefully (open Settings if denied)
My main challenges:
1.Detecting end of sentence reliably during continuous speech
2. Restarting recognition after each sentence for ongoing translation
3. Best way to handle auto language detection and free translation API
Any suggestions or working patterns would be great!