-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
246 lines (182 loc) · 9.53 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
<!doctype html>
<html style="width: 100%; height: 100%">
<head>
<meta charset="utf-8">
<title>wabps</title>
</head>
<body style="margin: 0; width: 100%; height: 100%; overflow: hidden">
<canvas id="output" width="1024" height="600" style="width: 100%; height: 50%; background: black; display: block"></canvas>
<canvas id="wa_output" width="1024" height="600" style="width: 100%; height: 50%; background: black"></canvas>
<button id="start_btn" style="position: absolute; left:0; top: 0;">Start</button>
<script type="text/javascript">
window.onload = async function () {
const getFrequency = (n, base_frequency, i, octaves) => {
return base_frequency * Math.pow(2, i / (n / octaves))
}
const canvas = document.getElementById("output")
const canvas_ctx = canvas.getContext('2d')
const tmp_canvas = document.getElementById("output")
const tmp_canvas_ctx = tmp_canvas.getContext('2d')
const wa_canvas = document.getElementById("wa_output")
const wa_canvas_ctx = wa_canvas.getContext("2d")
const tmp_wa_canvas = document.getElementById("wa_output")
const tmp_wa_canvas_ctx = wa_canvas.getContext("2d")
// create audio context
const audio_context = new window.AudioContext({
sampleRate: 44100 / 2
})
// = analysis parameters
// web audio bandpass filter bank
const minimum_frequency = 16.34
const maximum_frequency = audio_context.sampleRate / 2 // linear
const octaves = 10 // logarithmic
const frequency_step = 30 // linear
const analysis_window_size = 8192 / 2
const magnitude_factor = 128
// web audio analyzer node
const fft_size = analysis_window_size // NOTE : should be power of two for web audio analyzer (bp filter bank don't have that limitation)
const wa_magnitude_factor = 1
const wa_speed = 1
const wa_canvas_height = 100
// =
const speed = 1
canvas.width = 512
canvas.height = 100 // linear : Math.round((maximum_frequency - minimum_frequency) / frequency_step)
tmp_canvas.width = canvas.width
tmp_canvas.height = canvas.height
wa_canvas.width = canvas.width
wa_canvas.height = wa_canvas_height
tmp_wa_canvas.width = canvas.width
tmp_wa_canvas.height = wa_canvas_height
// = Web audio analyzer
const analyzer = audio_context.createAnalyser()
analyzer.fftSize = fft_size
analyzer.smoothingTimeConstant = 0
const analysis_data = new Uint8Array(analyzer.frequencyBinCount)
const draw = () => {
//requestAnimationFrame(draw)
analyzer.getByteFrequencyData(analysis_data)
tmp_wa_canvas_ctx.drawImage(tmp_wa_canvas, 0, 0, tmp_wa_canvas.width, tmp_wa_canvas.height)
for (let i = 0; i < wa_canvas.height; i += 1) {
const freq = getFrequency(wa_canvas.height, minimum_frequency, wa_canvas.height - i, octaves)
const bin_index = Math.round(freq * fft_size / audio_context.sampleRate)
const color = Math.round(analysis_data[bin_index] * wa_magnitude_factor)
wa_canvas_ctx.fillStyle = 'rgba(' + color + ',' + color + ',' + color + ',' + 1 + ')'
wa_canvas_ctx.fillRect(wa_canvas.width - 1 - wa_speed, i, wa_speed, wa_speed)
}
wa_canvas_ctx.translate(-wa_speed, 0)
wa_canvas_ctx.drawImage(tmp_wa_canvas, 0, 0, wa_canvas.width, wa_canvas.height, 0, 0, wa_canvas.width, wa_canvas.height)
wa_canvas_ctx.setTransform(1, 0, 0, 1, 0, 0)
}
// =
// compute blackman looktup table (avoid abrupt start & end, probably help improving quality; this is the same window used on Web Audio analyzer node for smoothingTimeConstant except Web Audio analyzer node apply it between buffers)
const a0 = 0.426591
const a1 = 0.496561
const a2 = 0.076849
const L = (analysis_window_size - 1)
const blackman_arr = new Float32Array(analysis_window_size)
for (let j = 0; j < analysis_window_size; j += 1) {
blackman_arr[j] = a0
blackman_arr[j] -= a1 * Math.cos((1.0 * 2.0 * Math.PI * j) / L)
blackman_arr[j] += a2 * Math.cos((2.0 * 2.0 * Math.PI * j) / L)
blackman_arr[j] -= 0.006879
}
audio_context.audioWorklet.addModule('capture-node.js').then(() => {
const capture_worklet_node = new AudioWorkletNode(audio_context, 'capture-node')
const audio_buffer = audio_context.createBuffer(1, analysis_window_size, audio_context.sampleRate)
let processing_duration = 0
let event_duration = 0
let last_event_time = performance.now()
capture_worklet_node.port.onmessage = (event) => {
const data = new Float32Array(event.data)
//audio_buffer.copyToChannel(data, 0)
const buffer_chn_data = audio_buffer.getChannelData(0)
for (let i = 0; i < buffer_chn_data.length; i += 1) {
buffer_chn_data[i] = data[i] * blackman_arr[i]
}
let last_processing_time = performance.now()
tmp_canvas_ctx.drawImage(canvas, 0, 0, canvas.width, canvas.height)
// do bandpass filter-bank analysis from worklet mic chunk data
for (let i = 0; i < canvas.height; i += 1) {
const offline_audio_ctx = new OfflineAudioContext(1, analysis_window_size, audio_context.sampleRate)
// create a bandpass filter
const bp_filter = offline_audio_ctx.createBiquadFilter()
bp_filter.type = 'bandpass'
bp_filter.connect(offline_audio_ctx.destination)
// create a second bandpass filter (24db rolloff)
const bp_filter2 = offline_audio_ctx.createBiquadFilter()
bp_filter2.type = 'bandpass'
bp_filter2.connect(bp_filter)
// create a second bandpass filter (48db rolloff)
const bp_filter3 = offline_audio_ctx.createBiquadFilter()
bp_filter3.type = 'bandpass'
bp_filter3.connect(bp_filter2)
// create a second bandpass filter (96db rolloff)
const bp_filter4 = offline_audio_ctx.createBiquadFilter()
bp_filter4.type = 'bandpass'
bp_filter4.connect(bp_filter3)
// run analysis for a single band
const from = getFrequency(canvas.height, minimum_frequency, canvas.height - i, octaves)
const to = getFrequency(canvas.height, minimum_frequency, canvas.height - i + 1, octaves) // linear : current_frequency + frequency_step
const geometric_mean = Math.sqrt(from * to)
// limit analysis up to nyquist
if (geometric_mean >= audio_context.sampleRate / 2) {
continue
}
bp_filter.frequency.value = geometric_mean
bp_filter.Q.value = geometric_mean / (to - from)
bp_filter2.frequency.value = geometric_mean
bp_filter2.Q.value = geometric_mean / (to - from)
bp_filter3.frequency.value = geometric_mean
bp_filter3.Q.value = geometric_mean / (to - from)
bp_filter4.frequency.value = geometric_mean
bp_filter4.Q.value = geometric_mean / (to - from)
const source_node = offline_audio_ctx.createBufferSource()
source_node.buffer = audio_buffer
source_node.connect(bp_filter) // replace it with bp_filter4 (96db) bp_filter3 (48db) bp_filter2 (24db) bp_filter (12db)
source_node.start(0)
// start frequency band analysis & rendering
offline_audio_ctx.startRendering().then((rendered_buffer) => {
const pcmf32_buffer = rendered_buffer.getChannelData(0)
// draw frequency band
let mean_amplitude = 0
for (let j = 0; j < pcmf32_buffer.length; j += 1) {
mean_amplitude += Math.abs(pcmf32_buffer[j])
}
mean_amplitude /= pcmf32_buffer.length
const color = Math.round(mean_amplitude * 255 * magnitude_factor);
canvas_ctx.fillStyle = 'rgba(' + color + ',' + color + ',' + color + ',' + 1 + ')'
canvas_ctx.fillRect(canvas.width - 1 - speed, i, speed, speed);
})
}
canvas_ctx.translate(-speed, 0)
canvas_ctx.drawImage(tmp_canvas, 0, 0, canvas.width, canvas.height, 0, 0, canvas.width, canvas.height)
canvas_ctx.setTransform(1, 0, 0, 1, 0, 0)
processing_duration = performance.now() - last_processing_time
event_duration = performance.now() - last_event_time
console.log("processing duration", processing_duration)
last_processing_time = performance.now()
console.log("event duration", event_duration)
last_event_time = performance.now()
draw()
};
const analysis_window_size_param = capture_worklet_node.parameters.get("analysisWindowSize")
analysis_window_size_param.setValueAtTime(analysis_window_size, audio_context.currentTime)
// microphone input
navigator.mediaDevices.getUserMedia({ audio: true, video: false })
.then((media_stream) => {
const input = audio_context.createMediaStreamSource(media_stream)
input.connect(capture_worklet_node)
input.connect(analyzer)
capture_worklet_node.connect(audio_context.destination)
//draw();
})
.catch((err) => { console.log(err.name + ": " + err.message) })
});
document.getElementById('start_btn').addEventListener('click', async () => {
await audio_context.resume()
})
}
</script>
</body>
</html>