Today I am excited to present two devices that will aid you in adding voice recognition services to your products. Speech recognition is becoming less of a novelty and more of a requirement in products and devices that are used in our daily lives. To help get you familiar with this technology we have a couple of kits that can get you on your way.
The Voice Capture Development Kit from Cirrus Logic (part number 598-2471-KIT) comes complete with the CRD1569-1 voice capture board, a Raspberry Pi, power supply, ribbon cable, and passive speaker. The voice capture board in this kit is essentially a HAT (Hardware Attached on Top) for the Raspberry Pi. This HAT gives you the ability to add Amazon Alexa voice services to your product. Additionally, the Alexa development software allows you to add custom vocabulary to your design.
The second product I want to discuss today comes from Seeed Technology in the form of the ReSpeaker Core which is based on MT7688 and OpenWRT (part number 102010088). Originally funded through Kickstarter, the ReSpeaker requires a bit more development than the Cirrus Kit I mentioned previously. This kit comes with the ReSpeaker Core and an 8 Gig Micro SD Card. The rest is up to you. There are a variety of boards that are compatible with the ReSpeaker, including the Grove System. This board runs OpenWRT which allows you to set up the ReSpeaker for network communication (as a router or switch for example). By utilizing the ATMega32U4 processor, you are able to program the ReSpeaker just as you would an Arduino device, or, if you wish, you can use the C, C++, or Python libraries. On top of that, there are open source projects available for offline speech recognition with the ability to tie this device to a PC and communicate with the Bing (Microsoft) word recognition system for speech-to-text and text-to-speech.
Have you used either of these devices? Let us know in the comments, and as always, thanks for watching!
