ai speech intent
The ai speech intent command recognizes streaming audio captured from a device microphone or stored in local or remote audio files and returns intent information from that recognition using PATTERNS.
Usage
Options
| Option | Description |
|---|---|
--key KEY |
Subscription key for the Azure Speech Service. |
--region REGION |
Region for the Azure Speech Service. |
--pattern PATTERN |
Simple pattern to use for recognizing intents. |
--pattern INTENTID=PATTERN |
Associate a pattern with a specific intent ID. |
--patterns PATTERN1;PATTERN2 |
Use multiple patterns for recognizing intents. |
--patterns @PATTERNS.txt |
Use patterns listed in a text file. |
--microphone |
Use the device microphone as the input source. |
--file FILE |
Use a local audio file as the input source. |
--files PATTERN |
Use multiple local audio files as input sources. |
--files @FILELIST.txt |
Use audio files listed in a text file as input sources. |
--format FORMAT |
Specify the audio format (e.g., wav, mp3). |
--once[+] |
Perform a single recognition for the input source (with optional repetition). |
--continuous |
Perform continuous recognition for the input source. |
--keyword FILENAME |
Perform recognition based on a keyword detected in the audio. |
--language LANGUAGE |
Specify the language of the input audio (e.g., en-US). |
--log FILENAME |
Save the recognition log to a file. |
--proxy HOSTNAME |
Use a proxy server for the recognition. |
--foreach in @ITEMS.txt |
Iterate the recognition over multiple items listed in a text file. |
--threads NUMBER |
Specify the number of threads to use for parallel processing. |
Examples
Recognize intent from a local audio file using a simple pattern
ai speech intent --file audio.wav --pattern "What's the weather like?"
Recognize intents from multiple audio files using patterns
ai speech intent --files *.wav --patterns "turn on the light";"turn off the light"