Getting - Started Route (Must - Read)
About 376 wordsAbout 1 min
2025-1-21
Whether you're a programmer or a novice, following the route below to use and learn ESP - AI
will ensure a smooth process.
Step 1: Fully Utilize Official Services
- Obtain an
ESP - AI development board
or prepare a set ofDIY kits
. See: Material Preparation. - Open the Open Platform and log in. For the first login, you need to use your mobile phone number.
- Create a super - body without changing any configurations.
- Finish reading the Open Platform usage tutorial: Firmware Burning, Network Configuration, and Basic Dialogue Tutorial
After completing the above steps, you'll have an AI assistant!
Step 2: Use Third - Party Services
When you find that the official ASR
, TTS
, and LLM
don't meet your requirements in terms of stability, quality, or speed, or if you want to experience the AI services of other platforms such as ByteDance, Alibaba, OpenAI, etc., you can register for services on these third - party platforms. For example, refer to the recommended third - party configurations: [Recommended Third - Party Configurations](/example/builtin-server/#recommended - third - party - configurations)
From this point on, you'll need to deal with the issues specific to third - party platforms, rather than simply using the product out - of - the - box. Each platform has its differences. For example, ByteDance requires you to activate voice tones and authorize your key before use, while iFLYTEK only requires you to activate the corresponding service...
If you can't even figure out how to register accounts on those platforms or activate services, it's best to stick with the open - platform services.
Step 3: Develop Your Own AI Application
If you're a programmer and can understand the development documentation, you can develop your own "Xiaoming Classmate" and use ESP - AI
to create any AI application.
For example, if you want to optimize the response speed to the extreme, you can set the client vad
to a few hundred milliseconds. For most scenarios, the default is set to 1500ms
. In short, when you develop an AI application based on ESP - AI
, you'll have a vast space for creativity.