108: Year of Voice: A Bigger Deal Than You Think
Self-Hosted
00:00
Google's Open Wake Word Model and ESP32 Limitations
This chapter discusses the use of an open wake word model created by Google for speech to text translation and the limitations it poses on the ESP32. The speakers express the desire for wake word detection and audio cleanup to occur directly on the ESP32 to prevent system overload. They also mention the option of disabling wake word listening in certain rooms using present sensors as a temporary solution.
Transcript
Play full episode
Transcript
Episode notes
Home Assistant's founder, Paulus Schoutsen, shares details about the Year of Voice, recent legal actions from Mazda, and the results of a recent third-party audit.
Plus, our recommended Nextcloud setup, converting dumb devices into smart ones with ESPHome, and more.
Special Guests: Brent Gervais, Listener Jeff, and Paulus Schoutsen.
Sponsored By:
- Linode: Receive a $100 60-day credit towards your new account. Promo Code: linode.com/ssh
- 45drives: The HL15 from 45HomeLab is an open-source, open-platform, 15-bay homelab server.
- Tailscale: Tailscale is a Zero config VPN. It installs on any device in minutes, manages firewall rules for you, and works from anywhere. Get 3 users and 100 devices for free.
Links:
- ⚡ Alby: Your Boost companion for the web — Create an Alby Account to get a lightning wallet for payments wherever you go.
- ⚡ Self-Hosted on the Podcastindex.org — Send a Boost into the show via the web. First, top-up Alby, then head over to our entry on the Podcast Index.
- ESPHome — ESPHome is a system to control your microcontrollers by simple yet powerful configuration files and control them remotely through Home Automation systems.
- WLED High Power Addressable LED Strip Controller — WLED Addressable Strip Controller
- 3pcs ESP8266 NodeMCU — ESP8266 CP2102 NodeMCU LUA ESP-12E WIFI Serial Wireless Module
- 2pcs 5V One Channel Relay Module Relay Switch — The module uses genuine quality relay, normally open interfaces , Maximum load: AC 250V/10A, DC 30V/10A.
- 2023: Home Assistant's year of Voice - Home Assistant — TL;DR: It is our goal for 2023 to let users control Home Assistant in their own language. Mike Hansen, creator of Rhasspy, has joined Nabu Casa to lead this effort. We’re starting off by building a collection of intent matching sentences in every
- Removal of Mazda Connected Services integration - Home Assistant — Home Assistant is disappointed that Mazda has decided to take this position. We’re also sad that Mazda’s first recourse was not to reach out to us and the maintainer but to send a cease and desist letter instead.
- Year of the Voice - Chapter 4 - YouTube — We’re building an open voice assistant that does not share your data. Join us for our fourth progress update. You don't want to miss this one.
- Year of the Voice - Chapter 4: Wake words - Home Assistant — We’ve got great news: wake words are finally here! After 4 chapters, we now have the final building block for voice in Home Assistant.
- $13 voice assistant for Home Assistant - Home Assistant — This tutorial will guide you to turn an ATOM Echo into the world’s most private voice assistant. Pick up the tiny device to talk to your smart home. Issue commands and get responses!
- ATOM Echo Smart Speaker Development Kit — ATOM ECHO is a Programmable Smart Speaker based on the M5ATOM design. With its compact form factor measuring only 24 * 24 * 17 mm, this speaker delivers impressive capabilities. Easily play music wirelessly using the BT capabilities of the ESP32 from your mobile phone or tablet.
- Unleash Your Creativity with ESP32-S3-BOX-3 — ESP32-S3-BOX-3 is based on Espressif’s ESP32-S3 Wi-Fi + Bluetooth 5 (LE) SoC, with AI acceleration capabilities. In addition to ESP32-S3’s 512KB SRAM, ESP32-S3-BOX-3 comes with 16MB of Quad flash and 16MB of Octal PSRAM.
- Anker PowerConf S330 USB Speakerphone — A USB speakerphone with 4 high-sensitivity microphones to pick up all voices within 3m in super-high clarity.
- piper: A fast, local neural text to speech system — A fast, local neural text to speech system that sounds great and is optimized for the Raspberry Pi 4. Piper is used in a variety of projects.
- wyoming: Peer-to-peer protocol for voice assistants — A peer-to-peer protocol for voice assistants (basically JSONL + PCM audio)
- wyoming-faster-whisper — Wyoming protocol server for faster whisper speech to text system
- openWakeWord: An open-source audio wake word (or phrase) detection framework with a focus on performance and simplicity. — openWakeWord is an open-source wakeword library that can be used to create voice-enabled applications and interfaces. It includes pre-trained models for common words & phrases that work well in real-world environments.
- Symfonium: Music player & cast - Apps on Google Play — Symfonium is a simple, modern, beautiful and highly configurable music player, with a unique vision to be the central control point of your music. Add one or multiple media providers.
- ZimaBoard - World's First Hackable Single Board Server — World’s First Hackable Single Board Server Starting at $119.9
The AI-powered Podcast Player
Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
