Creating conversational interfaces in iOS/Swift

on YouTube ↗20m

" summary

Compares Apple’s Speech framework (up to 1,000 requests per device per hour, 60s audio, iOS 10+, mostly online) with third‑party options like Amazon Lex (5,000 free requests per month first year) and Google (≈60 free minutes/month), shows a hands‑free recipe hack built with Lex using intents and slots, describes audio-in/audio-out vs text modes, and documents a practical hack to bypass Lex’s five retry limit by restarting listening; finishes with playback/testing notes and transparency/privacy rules.

AI-written summary · about

" chapters (7)

1. 0:00Start
2. 1:04Conversation as a Platform
3. 3:50iOS Speech Recognition API
4. 9:03Cost
5. 14:10Listen for input
6. 18:33Best Practices
7. 18:43Transparency!

" related talks

• 15mSwift Chatbots for Fun! and ProfitRay Tsaihong · try! Swift NYC 2017
• 01h 12mSpeeding Up with Accelerate: Experience Optimizing C++ CodeAleksey Alekseev · Mobius 2020 Piter
• 17mExploring Natural Language ProcessingPaola Mata · try! Swift NYC 2017
• 31mSwift, You, IAudrey Sobgou Zebaze · FrenchKit SE 2020
• 04mSponsors Lightning Interview #2MWM · FrenchKit 2019
• 25mAI-Assisted Swift in Gamifying an iOS AppChris Zai · try! Swift Tokyo 2025
• 19mDriving Engagement with Siri Shortcuts and NSUserActivityNic Laughter · try! Swift Tokyo 2019
• 26mSwift beyond Apple: Making Software for the Physical WorldJoannis Orlandos · Swift Rockies 2026