USB Camera, Microphone, MP3 Player and Text Speaker support for Tessel 2.
npm install tessel-av
The Camera class produces instances that may be used to capture a still frame or stream frames from a capable USB camera.
| Property | Type | Value/Description | Default | Required |
|---|---|---|---|---|
| device | string | The system path to the video device | /dev/video0 | no |
| dimensions | string | Valid "WxH" dimensions. Is limited to dimensions supported by the device.* | "800x600" | no |
| fps | number | Frames per second. Will be ignored if value is unsupported. | Per camera | no |
| port | number | Port number for the video server | 8080 | no |
| quality | number | Set the quality from 0...1 | 1 | no |
| fps | number | Frames per second. Will be ignored if value is unsupported. | Per camera | no |
* The device itself determines what dimensions are supported. Since the output comes directly from the camera hardware, invalid dimensions will be overridden by device if not supported.
capture Take a still frame. Returns a CaptureStream, call pipe with a destination to send a frame of jpeg encoded data.stream Stream mjpg frames from camera.data when stream has data.frame when camera has a complete frame.Use the data event to capture a single frame and save it as a JPEG:
'use strict';
const fs = require('fs');
const path = require('path');
const av = require('tessel-av');
const camera = new av.Camera();
const capture = camera.capture();
capture.on('data', function(data) {
fs.writeFile(path.join(__dirname, 'captures/captured-via-data-event.jpg'), data);
});
Respond to an HTTP request by piping the stream returned by capture():
'use strict';
const fs = require('fs');
const os = require('os');
const http = require('http');
const port = 8888;
const av = require('tessel-av');
const camera = new av.Camera({
width: 320,
height: 240,
});
const server = http.createServer((request, response) => {
response.writeHead(200, { "Content-Type": "image/jpg" });
camera.capture().pipe(response);
}).listen(port, () => console.log(`http://${os.hostname()}.local:${port}`));
process.on("SIGINT", _ => server.close());
A very simple example of direct-to-browser streaming:
'use strict';
const express = require('express');
const app = express();
const server = require('http').Server(app);
const os = require('os');
const path = require('path');
const port = 8888;
const av = require('tessel-av');
const camera = new av.Camera();
server.listen(port, function () {
console.log(`http://${os.hostname()}.local:${port}`);
});
app.use(express.static(path.join(__dirname, '/public')));
app.get('/stream', (request, response) => {
response.redirect(camera.url);
});
And here's the public/index.html file, which should be referenced in a .tesselinclude file):
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>streaming video to img element</title>
</head>
<body>
<img src="/stream">
</body>
</html>
The Player class produces instances that may be used to play audio from an mp3 file stored on the Tessel 2.
(Prior to v0.3.0, the av.Speaker class was used for audio file playback, and while that still works in versions >=0.3.0, programs will need to update to use this class before 1.0.0 (estimated release: July 1st, 2016)
The Player class constructor accepts one argument, which is optional, that specifies an mp3 file to play when the play() method is called. The may be omitted and supplied directly to play(file [, time]) at a later time in the object's lifecycle.
play([seconds]) Play the specified file. Optionally provide a time to start at in seconds. See Allowed Time String Formats
play([ file [, seconds]) Play the specified file. Optionally provide a time to start at in seconds. See Allowed Time String Formats
play([ 'file.mp3', ...options ]) Play the file with additional options in an array.
play(options) Play the file with additional options in an object.
Options {
phrase: "Hello Dave, you're looking well today",
...
}
pause() Pause playback of the current file.
stop() Stop playback of the current file (calling play() will start the playback from the beginning.)
| Format | Type |
|---|---|
hh:mm:ss | string |
ssss.dddd | number |
ssss | number |
Options may be most of options supported by madplay. For example, if I wanted to set the amplitude and pitch:
player.play(['foo.mp3', '-a', 10, '-r', 2 ]);
player.play(['foo.mp3', 'a', 10, 'r', 2 ]);
or
player.play({
file: 'foo.mp3',
a: 10,
r: 2,
});
ended when playback ends.play after play() is called.pause after pause() is called.stop after stop() is called.timeupdate approximately every 100ms. Delivers an approximation of the playback time in seconds, as ssss.ddd.The following is an example of the API and events working together:
'use strict';
const path = require('path');
const av = require('tessel-av');
const mp3 = path.join(__dirname, '20-second-nonsense.mp3');
const sound = new av.Player(mp3);
sound.play();
sound.on('timeupdate', function(seconds) {
seconds = Math.round(seconds);
if (seconds === 2) {
this.pause();
}
if (seconds > 12) {
this.stop().play();
}
});
sound.on('pause', function() {
setTimeout(() => this.play(10), 1000);
});
Alternatively, an mp3 file may be provided at the play(...) call site:
'use strict';
const path = require('path');
const av = require('./lib/index');
const mp3 = path.join(__dirname, 'some-audio-file.mp3');
const sound = new av.Player();
// Play the mp3, starting at the 10 second mark.
sound.play(mp3, "10");
sound.on('ended', function() {
console.log('This is not the end!');
sound.play();
});
Remember that you must explicitly specify static assets by listing them in a
.tesselincludefile file. For example, to ensure thesong.mp3file is deployed to your Tessel 2, you'll create a file called.tesselincludethat contains the following:song.mp3You may find it easier to put all static assets in a sub-directory, such as
public/, or similar. Then you can include all files and sub-directories by creating a.tesselincludefile and listing that assets directory like this:public
(Prior to v0.3.0, the av.Speaker class was used for audio file playback, and while that still works in versions >=0.3.0, programs will need to update to use this class before 1.0.0 (estimated release: July 1st, 2016)
say("phrase") Speak the phrase (string).say(["phrase", ...]) Speak the phrase with additional options in an array.say(options) Speak the phrase with additional options in an object.
Options {
phrase: "Hello Dave, you're looking well today",
...
}
stop() Stop playback.Options may be most of the options supported by espeak. For example, if I wanted to set the amplitude and pitch:
speaker.say(['Hello!', '-a', 10, '-p', 50 ]);
speaker.say(['Hello!', 'a', 10, 'p', 50 ]);
or
speaker.say({
phrase: 'Hello!',
a: 10,
p: 50,
});
Back to back calls to speaker.say(...) will result in each phrase being queued. Once a phrase has been said, the next phrase in the queue will be spoken.
ended when speech ends.empty when speech ends and the speech queue is empty.lastword when speech ends and the speech queue is empty and the program should prevent any further empty events from being emitted. This allows your robot to get the last word in without repeating themselves forever.say after say() is called.stop after stop() is called.timeupdate approximately every 100ms. Delivers an approximation of the speech time in seconds, as ssss.ddd.The following is an example of the API and events working together:
ended event will trigger, which will start the "cycle" through the letters of the alphabet.'use strict';
const os = require('os');
const path = require('path');
const av = require('tessel-av');
const alphabet = 'abcdefghijklmnopqrstuvwxyz'.split('');
const speaker = new av.Speaker();
speaker.say(`
Hello, this is ${os.hostname()}.
I'm going to say my A-B-C's now
`);
speaker.on('ended', function() {
if (alphabet.length) {
this.say(alphabet.shift());
}
});
Alternatively, each letter can be "queued":
'use strict';
const os = require('os');
const av = require('tessel-av');
const alphabet = 'abcdefghijklmnopqrstuvwxyz'.split('');
const speaker = new av.Speaker();
speaker.say(`
Hello, this is ${os.hostname()}.
I'm going to say my A-B-C's now
`);
alphabet.forEach(letter => speaker.say(letter));
speaker.on('lastword', function() {
// If this had been an `empty` event, it would've
// been emitted again as soon as the next phrase
// was spoken.
this.say('And now I know my A-B-Cs');
});
MIT.