// Copyright 2021 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
GEN_INCLUDE(['enhanced_network_tts_e2e_test_base.js']);
/**
* Test fixture for enhanced_network_tts_unittest.js.
*/
EnhancedNetworkTtsUnitTest = class extends EnhancedNetworkTE2ETestBase {
/** @override */
testGenPreamble() {
super.testGenPreamble();
super.testGenPreambleCommon(
'kEnhancedNetworkTtsExtensionId', true /* failOnConsoleError */);
}
};
SYNC_TEST_F(
'EnhancedNetworkTtsUnitTest', 'onSpeakWithAudioStreamEventSucceed',
async function() {
// Prepare the mockTtsApi to respond with audio data that corresponds to a
// 0.2135s playback containing "Hello world".
const mockTtsApi = MockTtsApi;
mockTtsApi.enqueueAudioData(
generateTestBufferData(), generateTestTimeInfoData(),
true /* lastData */);
chrome.mojoPrivate.registerMockedModuleForTesting(
'ash.enhanced_network_tts', mockTtsApi);
const utterance = 'test';
const options = {'voiceName': 'Enhanced TTS English (Australian Accent)'};
const sampleRate = 10000;
const bufferSize = 400;
const audioStreamOptions = {bufferSize, sampleRate};
const decodedAudioData =
await EnhancedNetworkTts.decodeAudioDataAtSampleRate(
generateTestBufferData(), sampleRate);
// Each buffer corresponds to 0.04s.
const expectedBuffers = [
// The first 0.04s is empty.
{
audioBuffer:
EnhancedNetworkTts.subarrayFrom(decodedAudioData, 0, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// 0.04s - 0.08s contains the start of the first word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize, bufferSize),
charIndex: 0,
isLastBuffer: false,
},
// 0.08s - 0.12s is a part of the first word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 2, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// 0.12s - 0.16s contains the end of the first word and the start of the
// second.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 3, bufferSize),
charIndex: 6,
isLastBuffer: false,
},
// 0.16s - 0.2s is a part of the second word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 4, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// The last buffer runs 0.0135s.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 5, bufferSize),
charIndex: undefined,
isLastBuffer: true,
},
];
const sendTtsAudio = function(receivedBuffer) {
const expectedBuffer = expectedBuffers.shift();
assertEqualsJSON(expectedBuffer, receivedBuffer);
};
await enhancedNetworkTts.onSpeakWithAudioStreamEvent(
utterance, options, audioStreamOptions, sendTtsAudio);
});
SYNC_TEST_F('EnhancedNetworkTtsUnitTest', 'GenerateRequest', async function() {
let utterance = 'name and lang should be specified together';
let options = {voiceName: 'test name', lang: 'en'};
let request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 1.0, voice: 'test name', lang: 'en'});
utterance = 'name without lang will be ignored';
options = {voiceName: 'test name'};
request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 1.0, voice: undefined, lang: undefined});
utterance = 'lang without name will be ignored';
options = {lang: 'en'};
request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 1.0, voice: undefined, lang: undefined});
utterance = 'only lang code (e.g., en) will be used';
options = {voiceName: 'test name', lang: 'en_US'};
request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 1.0, voice: 'test name', lang: 'en'});
utterance = 'utterance without options can proceed';
options = {};
request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 1.0, voice: undefined, lang: undefined});
utterance = 'Rate will be sent along with the request';
options = {rate: 3.0};
request = EnhancedNetworkTts.generateRequest(utterance, options);
assertEqualsJSON(
request, {utterance, rate: 3.0, voice: undefined, lang: undefined});
});
SYNC_TEST_F(
'EnhancedNetworkTtsUnitTest', 'DecodeAudioDataAtSampleRate',
async function() {
const testAudioLength = 0.2135;
let sampleRate = 4000;
let audioBuffer = await EnhancedNetworkTts.decodeAudioDataAtSampleRate(
generateTestBufferData(), sampleRate);
assertEquals(
audioBuffer.length, Math.floor(testAudioLength * sampleRate));
sampleRate = 6000;
audioBuffer = await EnhancedNetworkTts.decodeAudioDataAtSampleRate(
generateTestBufferData(), sampleRate);
assertEquals(
audioBuffer.length, Math.floor(testAudioLength * sampleRate));
sampleRate = 10000;
audioBuffer = await EnhancedNetworkTts.decodeAudioDataAtSampleRate(
generateTestBufferData(), sampleRate);
assertEquals(
audioBuffer.length, Math.floor(testAudioLength * sampleRate));
});
SYNC_TEST_F('EnhancedNetworkTtsUnitTest', 'SubarrayFrom', async function() {
const sampleArray = new Float32Array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10]);
let subarray = EnhancedNetworkTts.subarrayFrom(
sampleArray, 0 /* startIdx */, 5 /* subarraySize */);
assertArraysEquals(subarray, new Float32Array([1, 2, 3, 4, 5]));
subarray = EnhancedNetworkTts.subarrayFrom(
sampleArray, 5 /* startIdx */, 5 /* subarraySize */);
assertArraysEquals(subarray, new Float32Array([6, 7, 8, 9, 10]));
subarray = EnhancedNetworkTts.subarrayFrom(
sampleArray, 5 /* startIdx */, 10 /* subarraySize */);
assertArraysEquals(
subarray, new Float32Array([6, 7, 8, 9, 10, 0, 0, 0, 0, 0]));
subarray = EnhancedNetworkTts.subarrayFrom(
sampleArray, 11 /* startIdx */, 10 /* subarraySize */);
assertArraysEquals(
subarray, new Float32Array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0]));
});
SYNC_TEST_F(
'EnhancedNetworkTtsUnitTest', 'SendAudioDataInBuffers', async function() {
// Prepare the decodedAudioData for testing. The data corresponds to a
// 0.2135s playback.
const testAudioLength = 0.2135;
const sampleRate = 10000;
const decodedAudioDataLength = testAudioLength * sampleRate; // 2135
const decodedAudioData =
await EnhancedNetworkTts.decodeAudioDataAtSampleRate(
generateTestBufferData(), sampleRate);
assertEquals(decodedAudioData.length, decodedAudioDataLength);
const timeInfo = generateTestTimeInfoData();
// The |decodedAudioData| will be sent through 6 buffers. The first five
// buffers has a size of 200, and the last one only has a size of 135.
const bufferSize = 400;
// Each buffers corresponds to 0.04s in the playback.
const expectedBuffers = [
// The first 0.04s is empty.
{
audioBuffer:
EnhancedNetworkTts.subarrayFrom(decodedAudioData, 0, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// 0.04s - 0.08s contains the start of the first word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize, bufferSize),
charIndex: 0,
isLastBuffer: false,
},
// 0.08s - 0.12s is a part of the first word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 2, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// 0.12s - 0.16s contains the end of the first word and the start of the
// second.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 3, bufferSize),
charIndex: 6,
isLastBuffer: false,
},
// 0.16s - 0.2s is a part of the second word.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 4, bufferSize),
charIndex: undefined,
isLastBuffer: false,
},
// The last buffer runs 0.0135s.
{
audioBuffer: EnhancedNetworkTts.subarrayFrom(
decodedAudioData, bufferSize * 5, bufferSize),
charIndex: undefined,
isLastBuffer: true,
},
];
// Copy the expectedBuffers but modify |isLastBuffer|.
const expectedBuffersWithoutLastBuffer =
expectedBuffers.map(buffer => Object.assign({}, buffer));
expectedBuffersWithoutLastBuffer[5].isLastBuffer = false;
let mockSendTtsAudio = receivedBuffer => {
const expectedBuffer = expectedBuffers.shift();
assertEqualsJSON(expectedBuffer, receivedBuffer);
};
EnhancedNetworkTts.sendAudioDataInBuffers(
decodedAudioData, sampleRate, bufferSize, timeInfo, mockSendTtsAudio,
/* lastData= */ true);
mockSendTtsAudio = receivedBuffer => {
const expectedBuffer = expectedBuffersWithoutLastBuffer.shift();
assertEqualsJSON(expectedBuffer, receivedBuffer);
};
// Will not signal the last buffer if |lastData| is false.
EnhancedNetworkTts.sendAudioDataInBuffers(
decodedAudioData, sampleRate, bufferSize, timeInfo, mockSendTtsAudio,
/* lastData= */ false);
});
/**
* Generates audio data for testing. The audio data contains a clip of silent
* audio originally in ogg format. The audio lasts 0.2135 seconds.
* @return {!Array<number>}
*/
function generateTestBufferData() {
const testData = [
79, 103, 103, 83, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 138,
1, 148, 99, 0, 0, 0, 0, 19, 72, 225, 168, 1, 19, 79, 112,
117, 115, 72, 101, 97, 100, 1, 1, 56, 1, 192, 93, 0, 0, 0,
0, 0, 79, 103, 103, 83, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 138, 1, 148, 99, 1, 0, 0, 0, 225, 155, 98, 95, 1, 40,
79, 112, 117, 115, 84, 97, 103, 115, 6, 0, 0, 0, 71, 111, 111,
103, 108, 101, 1, 0, 0, 0, 14, 0, 0, 0, 101, 110, 99, 111,
100, 101, 114, 61, 71, 111, 111, 103, 108, 101, 79, 103, 103, 83, 0,
4, 208, 41, 0, 0, 0, 0, 0, 0, 138, 1, 148, 99, 2, 0,
0, 0, 89, 72, 23, 6, 11, 14, 15, 16, 16, 16, 15, 15, 16,
26, 20, 22, 104, 11, 228, 193, 34, 35, 97, 249, 140, 57, 108, 66,
108, 208, 104, 7, 201, 121, 197, 18, 247, 188, 172, 131, 121, 219, 241,
240, 200, 104, 7, 201, 121, 197, 19, 42, 199, 60, 74, 211, 193, 207,
154, 151, 104, 104, 7, 201, 114, 39, 225, 62, 83, 8, 232, 252, 15,
209, 23, 151, 104, 104, 7, 201, 114, 39, 225, 62, 83, 8, 232, 252,
15, 209, 23, 151, 104, 104, 7, 201, 121, 197, 18, 247, 188, 172, 131,
121, 250, 81, 240, 200, 104, 7, 201, 121, 197, 18, 247, 188, 172, 131,
121, 250, 81, 240, 200, 104, 7, 201, 114, 39, 225, 62, 83, 8, 232,
252, 15, 209, 23, 151, 104, 104, 0, 48, 103, 26, 34, 194, 86, 230,
218, 164, 141, 181, 7, 252, 252, 185, 183, 181, 96, 71, 167, 143, 221,
7, 142, 104, 12, 209, 169, 168, 188, 105, 184, 244, 111, 145, 71, 40,
74, 40, 33, 234, 158, 16, 44, 104, 9, 73, 154, 65, 180, 184, 46,
212, 58, 33, 41, 158, 252, 16, 100, 140, 106, 65, 21, 168, 221,
];
return new Uint8Array(testData).buffer;
}
/**
* Generates timestamp data for two words, "Hello world". The first 0.05s in the
* playback is empty. "Hello" is spoken during 0.05 - 0.15s, "world" is spoken
* during 0.15 - 0.21s. The last 0.0035s is empty, assuming the audio lasts
* 0.2135 seconds.
* @return {!Array<!ash.enhancedNetworkTts.mojom.TimingInfo>}
*/
function generateTestTimeInfoData() {
return [
// The first 0.05s in the playback is empty.
{
'text': 'Hello',
'textOffset': 0, // The word offset to the start of the utterance.
'timeOffset': '0.05s', // Start time at the entire playback.
'duration': '0.1s', // The word lasts 0.1s.
},
{
'text': 'world',
'textOffset': 6, // The word offset to the start of the utterance.
'timeOffset': '0.15s', // Start time at the entire playback.
'duration': '0.06s', // The word lasts 0.6s.
}, // The last 0.0035s in the playback is empty.
];
}
Codebase Browser
chromium