February 2023 Summaries

This guide demonstrates how to build a sample backend app in Node.js that uses 100ms REST APIs for creating rooms, generating Auth Tokens, and getting usage analytics for the latest session in the room. The application consists of two components: client (frontend) and server (backend). The server communicates with 100ms REST APIs to perform various operations. Key terminologies like Room, Auth Token, and Session are explained. The guide provides step-by-step instructions for setting up the environment, creating services for token management and API calls, and implementing request handlers for different endpoints. Finally, it suggests deploying the app on Render for free.

A customer requested a feature for a social audio rooms app on Android, which involved transforming certain rooms into "radio stations" by allowing hosts to stream local music while speaking. Since WebRTC on mobile doesn't support adding custom audio tracks, significant modifications were made to the WebRTC source and Android SDK to enable mixing audio captured from the device's microphone with local audio. The process involved capturing microphone audio using WebRTC's JavaAudioDeviceModule and capturing local device audio with Android’s AudioRecord class, which required Android 10 or above. A callback mechanism was employed to synchronize the capture of audio bytes from both sources, which were then mixed programmatically to create a combined audio stream broadcasted to listeners. This solution also provides the flexibility to replace microphone audio with local audio if needed. The changes aim to enhance the app's functionality, and developers are encouraged to share feedback on the implementation through Discord.

The text discusses the implementation of a feature for an Android social audio rooms app, where users can host talk shows and stream local music simultaneously. To achieve this, significant changes were made to WebRTC and the Android SDK. The problem was broken down into three parts: capturing audio from the device mic, capturing local audio from the device, and mixing two audio streams. The first part involved understanding how audio is captured by WebRTC and creating a callback to get the byte buffer read from the microphone. In the second part, Android's AudioRecord class was used to record system audio played by other apps on the device. The third part focused on mixing the two audio streams using a method that combines the ShortArrays of both streams into a single ByteArray. The final solution involves capturing audio from both sources and replacing the microphone byte buffer with the resultant combined audio stream, allowing users to host talk shows while streaming local music simultaneously.

100ms, a real-time video conferencing SDK, has updated its pricing model to offer free minutes for all capabilities including interactive live streaming, external streaming, and recording. Starting from the January 2023 billing cycle, developers can use 10,000 free minutes for real-time video conferencing, 10,000 viewer minutes and 1,000 encoding minutes for interactive live streaming, 300 minutes for external streaming, and 300 minutes for recording. Additionally, new features have been added to the 100ms Dashboard for a more self-serve billing experience, including a complete usage overview, review and payment of invoices, and adding a payment method using Stripe.