mozilla
/
DeepSpeech-examples
зеркало из https://github.com/mozilla/DeepSpeech-examples.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

Bump version from 0.8.0 to 0.9.3 in branch r0.9 

Bump DeepSpeech version from 0.8.0 to 0.9.3 in branch r0.9
This commit is contained in:
Hadran Ayoub 2021-01-06 18:48:06 +01:00
Родитель 2b6ca883c0
Коммит 49b9832187
32 изменённых файлов: 66 добавлений и 66 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

8
android_mic_streaming/README.md
Просмотреть файл

@ -8,15 +8,15 @@ Android demo application that streams audio from the microphone to deepspeech an
Download the pre-trained English model and extract it: Download the pre-trained English model and extract it:
``` ```
curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.tflite curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.tflite
curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer curl -LO https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```
Move the model files `deepspeech-0.8.0-models.pbmm`, `deepspeech-0.8.0-models.scorer`, to the demo application's data directory on your android device. Move the model files `deepspeech-0.9.3-models.pbmm`, `deepspeech-0.9.3-models.scorer`, to the demo application's data directory on your android device.
Mind that the data directory will only be present after installing and launching the app once. Mind that the data directory will only be present after installing and launching the app once.
``` ```
adb push deepspeech-0.8.0-models.tflite deepspeech-0.8.0-models.scorer /storage/emulated/0/Android/data/org.deepspeechdemo/files/ adb push deepspeech-0.9.3-models.tflite deepspeech-0.9.3-models.scorer /storage/emulated/0/Android/data/org.deepspeechdemo/files/
``` ```
You can also copy the files from your file browser to the device. You can also copy the files from your file browser to the device.

2
android_mic_streaming/app/build.gradle
Просмотреть файл

@ -34,7 +34,7 @@ dependencies {
implementation 'androidx.core:core-ktx:1.0.2' implementation 'androidx.core:core-ktx:1.0.2'
implementation 'androidx.constraintlayout:constraintlayout:1.1.3' implementation 'androidx.constraintlayout:constraintlayout:1.1.3'
implementation 'org.mozilla.deepspeech:libdeepspeech:0.8.0' implementation 'org.mozilla.deepspeech:libdeepspeech:0.9.3'
testImplementation 'junit:junit:4.12' testImplementation 'junit:junit:4.12'
androidTestImplementation 'androidx.test.ext:junit:1.1.0' androidTestImplementation 'androidx.test.ext:junit:1.1.0'

4
android_mic_streaming/app/src/main/java/org/deepspeechdemo/MainActivity.kt
Просмотреть файл

@ -21,8 +21,8 @@ class MainActivity : AppCompatActivity() {
private var transcriptionThread: Thread? = null private var transcriptionThread: Thread? = null
private var isRecording: AtomicBoolean = AtomicBoolean(false) private var isRecording: AtomicBoolean = AtomicBoolean(false)
private val TFLITE_MODEL_FILENAME = "deepspeech-0.8.0-models.tflite" private val TFLITE_MODEL_FILENAME = "deepspeech-0.9.3-models.tflite"
private val SCORER_FILENAME = "deepspeech-0.8.0-models.scorer" private val SCORER_FILENAME = "deepspeech-0.9.3-models.scorer"
private fun checkAudioPermission() { private fun checkAudioPermission() {
// Permission is automatically granted on SDK < 23 upon installation. // Permission is automatically granted on SDK < 23 upon installation.

4
batch_processing/Readme.md
Просмотреть файл

@ -128,13 +128,13 @@ Running via the GPU takes half the time of using the CPU and has good results.
# Driver command line # Driver command line
`./driver.py --model c:/Users/jmike/Documents/GitHub/DeepSpeech/deepspeech-0.8.0-models.pbmm --scorer c:/Users/jmike/Documents/GitHub/DeepSpeech/deepspeech-0.8.0-models.scorer --dirname c:/Users/jmike/Downloads/podcast/` `./driver.py --model c:/Users/jmike/Documents/GitHub/DeepSpeech/deepspeech-0.9.3-models.pbmm --scorer c:/Users/jmike/Documents/GitHub/DeepSpeech/deepspeech-0.9.3-models.scorer --dirname c:/Users/jmike/Downloads/podcast/`
# Example # Example
It will then run the individual commands like : It will then run the individual commands like :
`deepspeech --model C:\Users\jmike\Documents\GitHub\DeepSpeech\deepspeech-0.8.0-models.pbmm --scorer C:\Users\jmike\Documents\GitHub\DeepSpeech\deepspeech-0.8.0-models.scorer --audio 'C:\Users\jmike\Downloads\podcast\45374977-48000-2-24d9a365625bb.mp3.wav' --json` `deepspeech --model C:\Users\jmike\Documents\GitHub\DeepSpeech\deepspeech-0.9.3-models.pbmm --scorer C:\Users\jmike\Documents\GitHub\DeepSpeech\deepspeech-0.9.3-models.scorer --audio 'C:\Users\jmike\Downloads\podcast\45374977-48000-2-24d9a365625bb.mp3.wav' --json`
Websites referenced: Websites referenced:

4
batch_processing/driver.py
Просмотреть файл

@ -19,12 +19,12 @@ import delegator
@click.option("--ext", default=".mp3") @click.option("--ext", default=".mp3")
@click.option( @click.option(
"--model", "--model",
default="deepspeech-0.8.0-models.pbmm", default="deepspeech-0.9.3-models.pbmm",
type=click.Path(exists=True, resolve_path=True), type=click.Path(exists=True, resolve_path=True),
) )
@click.option( @click.option(
"--scorer", "--scorer",
default="deepspeech-0.8.0-models.scorer", default="deepspeech-0.9.3-models.scorer",
type=click.Path(exists=True, resolve_path=True), type=click.Path(exists=True, resolve_path=True),
) )

2
batch_processing/requirements.txt
Просмотреть файл

@ -10,7 +10,7 @@ cachetools==4.1.0
certifi==2020.4.5.2 certifi==2020.4.5.2
chardet==3.0.4 chardet==3.0.4
click==7.1.2 click==7.1.2
deepspeech==0.8.0 deepspeech==0.9.3
delegator.py @ git+https://github.com/amitt001/delegator.py.git@194aa92543fbdbfbae0bcc24ca217819a7805da2 delegator.py @ git+https://github.com/amitt001/delegator.py.git@194aa92543fbdbfbae0bcc24ca217819a7805da2
flask==1.1.2 flask==1.1.2
gast==0.2.2 gast==0.2.2

10
electron/Readme.md
Просмотреть файл

@ -14,17 +14,17 @@ npm run rebuild
Download and extract audio files to `/public` directory Download and extract audio files to `/public` directory
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/audio-0.8.0.tar.gz wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/audio-0.9.3.tar.gz
tar xfvz audio-0.8.0.tar.gz -C ./public/ tar xfvz audio-0.9.3.tar.gz -C ./public/
``` ```
(Optional) Download or softlink DeepSpeech 0.8.0 model files to the root of the project: (Optional) Download or softlink DeepSpeech 0.9.3 model files to the root of the project:
``` ```
mkdir models mkdir models
cd models cd models
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
cd .. cd ..
``` ```

4
electron/package-lock.json сгенерированный
Просмотреть файл

@ -4791,8 +4791,8 @@
"integrity": "sha1-s2nW+128E+7PUk+RsHD+7cNXzzQ=" "integrity": "sha1-s2nW+128E+7PUk+RsHD+7cNXzzQ="
}, },
"deepspeech": { "deepspeech": {
"version": "0.8.0", "version": "0.9.3",
"resolved": "https://registry.npmjs.org/deepspeech/-/deepspeech-0.8.0.tgz", "resolved": "https://registry.npmjs.org/deepspeech/-/deepspeech-0.9.3.tgz",
"integrity": "sha512-jqU+NbXVZnS+okMgoiOhJz22RaHSmvIjmHaRu7IZ0xBDQbcqNGff4GXk4a5etfSXm3bXddRtBlfFr5KyQExjbw==", "integrity": "sha512-jqU+NbXVZnS+okMgoiOhJz22RaHSmvIjmHaRu7IZ0xBDQbcqNGff4GXk4a5etfSXm3bXddRtBlfFr5KyQExjbw==",
"requires": { "requires": {
"argparse": "1.0.x", "argparse": "1.0.x",

2
electron/package.json
Просмотреть файл

@ -66,7 +66,7 @@
"keywords": [], "keywords": [],
"license": "MIT", "license": "MIT",
"dependencies": { "dependencies": {
"deepspeech": "^0.8.0", "deepspeech": "^0.9.3",
"electron-is-dev": "^1.1.0", "electron-is-dev": "^1.1.0",
"lodash": "^4.17.15", "lodash": "^4.17.15",
"node-abi": "^2.18.0", "node-abi": "^2.18.0",

2
electron/public/electron.js
Просмотреть файл

@ -7,7 +7,7 @@ const {getModel} = require('./recognize-wav');
let appDataPath; let appDataPath;
if (fs.existsSync(path.resolve(__dirname, '../models/deepspeech-0.8.0-models.pbmm'))) { if (fs.existsSync(path.resolve(__dirname, '../models/deepspeech-0.9.3-models.pbmm'))) {
// if the deepspeech model was found at the root, use that directory // if the deepspeech model was found at the root, use that directory
appDataPath = path.resolve(__dirname, '../models'); appDataPath = path.resolve(__dirname, '../models');
} }

6
electron/public/recognize-wav.js
Просмотреть файл

@ -6,15 +6,15 @@ const download = require('./download');
// return the deepspeech model or download it if it is not found // return the deepspeech model or download it if it is not found
function getModel(appDataPath, callback) { function getModel(appDataPath, callback) {
let modelPath = path.resolve(appDataPath, 'deepspeech-0.8.0-models.pbmm'); let modelPath = path.resolve(appDataPath, 'deepspeech-0.9.3-models.pbmm');
let scorerPath = path.resolve(appDataPath, 'deepspeech-0.8.0-models.scorer'); let scorerPath = path.resolve(appDataPath, 'deepspeech-0.9.3-models.scorer');
if (fs.existsSync(modelPath) && fs.existsSync(scorerPath)) { if (fs.existsSync(modelPath) && fs.existsSync(scorerPath)) {
callback(createModel(modelPath, scorerPath)); callback(createModel(modelPath, scorerPath));
} }
else { else {
// if the model files do not exist, download and save them to AppData path // if the model files do not exist, download and save them to AppData path
console.log('\nDOWNLOADING MODEL TO: '+appDataPath+'\n'); console.log('\nDOWNLOADING MODEL TO: '+appDataPath+'\n');
const downloadPath = 'https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models'; const downloadPath = 'https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models';
download(downloadPath+'.pbmm', modelPath, function() { download(downloadPath+'.pbmm', modelPath, function() {
download(downloadPath+'.scorer', scorerPath, function() { download(downloadPath+'.scorer', scorerPath, function() {
callback(createModel(modelPath, scorerPath)); callback(createModel(modelPath, scorerPath));

2
ffmpeg_vad_streaming/package.json
Просмотреть файл

@ -8,7 +8,7 @@
}, },
"dependencies": { "dependencies": {
"argparse": "^1.0.10", "argparse": "^1.0.10",
"deepspeech": "0.8.0", "deepspeech": "0.9.3",
"node-vad": "^1.1.1", "node-vad": "^1.1.1",
"util": "^0.11.1" "util": "^0.11.1"
}, },

12
ffmpeg_vad_streaming/test.sh
Просмотреть файл

@ -11,14 +11,14 @@ pushd ${THIS}
npm install npm install
node ./index.js --audio $HOME/DeepSpeech/audio/2830-3980-0043.wav \ node ./index.js --audio $HOME/DeepSpeech/audio/2830-3980-0043.wav \
--scorer $HOME/DeepSpeech/models/deepspeech-0.8.0-models.scorer \ --scorer $HOME/DeepSpeech/models/deepspeech-0.9.3-models.scorer \
--model $HOME/DeepSpeech/models/deepspeech-0.8.0-models.pbmm --model $HOME/DeepSpeech/models/deepspeech-0.9.3-models.pbmm
node ./index.js --audio $HOME/DeepSpeech/audio/4507-16021-0012.wav \ node ./index.js --audio $HOME/DeepSpeech/audio/4507-16021-0012.wav \
--scorer $HOME/DeepSpeech/models/deepspeech-0.8.0-models.scorer \ --scorer $HOME/DeepSpeech/models/deepspeech-0.9.3-models.scorer \
--model $HOME/DeepSpeech/models/deepspeech-0.8.0-models.pbmm --model $HOME/DeepSpeech/models/deepspeech-0.9.3-models.pbmm
node ./index.js --audio $HOME/DeepSpeech/audio/8455-210777-0068.wav \ node ./index.js --audio $HOME/DeepSpeech/audio/8455-210777-0068.wav \
--scorer $HOME/DeepSpeech/models/deepspeech-0.8.0-models.scorer \ --scorer $HOME/DeepSpeech/models/deepspeech-0.9.3-models.scorer \
--model $HOME/DeepSpeech/models/deepspeech-0.8.0-models.pbmm --model $HOME/DeepSpeech/models/deepspeech-0.9.3-models.pbmm
popd popd

2
mic_vad_streaming/requirements.txt
Просмотреть файл

@ -1,4 +1,4 @@
deepspeech~=0.8.0 deepspeech~=0.9.3
pyaudio~=0.2.11 pyaudio~=0.2.11
webrtcvad~=2.0.10 webrtcvad~=2.0.10
halo~=0.0.18 halo~=0.0.18

4
mic_vad_streaming/test.sh
Просмотреть файл

@ -13,7 +13,7 @@ pushd ${THIS}
pulseaudio & pulseaudio &
python mic_vad_streaming.py \ python mic_vad_streaming.py \
--model $HOME/DeepSpeech/models/deepspeech-0.8.0-models.pbmm \ --model $HOME/DeepSpeech/models/deepspeech-0.9.3-models.pbmm \
--scorer $HOME/DeepSpeech/models/deepspeech-0.8.0-models.scorer \ --scorer $HOME/DeepSpeech/models/deepspeech-0.9.3-models.scorer \
--file $HOME/DeepSpeech/audio/2830-3980-0043.wav --file $HOME/DeepSpeech/audio/2830-3980-0043.wav
popd popd

2
net_framework/DeepSpeechWPF/App.xaml.cs
Просмотреть файл

@ -20,7 +20,7 @@ namespace DeepSpeechWPF
{ {
//Register instance of DeepSpeech //Register instance of DeepSpeech
DeepSpeechClient.DeepSpeech deepSpeechClient = DeepSpeechClient.DeepSpeech deepSpeechClient =
new DeepSpeechClient.DeepSpeech("deepspeech-0.8.0-models.pbmm"); new DeepSpeechClient.DeepSpeech("deepspeech-0.9.3-models.pbmm");
SimpleIoc.Default.Register<IDeepSpeech>(() => deepSpeechClient); SimpleIoc.Default.Register<IDeepSpeech>(() => deepSpeechClient);
SimpleIoc.Default.Register<MainWindowViewModel>(); SimpleIoc.Default.Register<MainWindowViewModel>();

2
nim_mic_vad_streaming/README.md
Просмотреть файл

@ -10,7 +10,7 @@ Interface to both the libs is provided through NIM code.
## PREREQUISITIES : ## PREREQUISITIES :
* ```libdeepspeech.so``` * ```libdeepspeech.so```
Go to the [releases](https://github.com/mozilla/DeepSpeech/releases/tag/v0.8.0) page and download the native client package based on your OS and CPU architecture. Go to the [releases](https://github.com/mozilla/DeepSpeech/releases/tag/v0.9.3) page and download the native client package based on your OS and CPU architecture.
Extract the ``libdeepspeech.so`` and put into the subdirectory depending on OS of native Client used. Extract the ``libdeepspeech.so`` and put into the subdirectory depending on OS of native Client used.

4
nim_mic_vad_streaming/linux_nim_vad_streaming/README.md
Просмотреть файл

@ -12,8 +12,8 @@ sudo apt-get install libasound2
2) Download the pre-trained DeepSpeech english model (1089MB) and Scorer Package(~900MB): 2) Download the pre-trained DeepSpeech english model (1089MB) and Scorer Package(~900MB):
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```

4
nim_mic_vad_streaming/win_nim_vad_streaming/README.md
Просмотреть файл

@ -8,8 +8,8 @@ Minimalistic example to demonstrate the DeepSpeech streaming API in NIM.Raw aud
2) Download the pre-trained DeepSpeech english model (1089MB): 2) Download the pre-trained DeepSpeech english model (1089MB):
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```

6
nodejs_mic_vad_streaming/Readme.md
Просмотреть файл

@ -11,8 +11,8 @@ either [sox](http://sox.sourceforge.net/) (Windows/Mac) or [arecord](http://alsa
2) Download the pre-trained DeepSpeech english model (1089MB): 2) Download the pre-trained DeepSpeech english model (1089MB):
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```
#### Dependency #### Dependency
@ -40,5 +40,5 @@ node start.js
Use the `DEEPSPEECH_MODEL` environment variable to change models. Use the `DEEPSPEECH_MODEL` environment variable to change models.
``` ```
DEEPSPEECH_MODEL=~/dev/jaxcore/deepspeech-0.8.0-models/ node start.js DEEPSPEECH_MODEL=~/dev/jaxcore/deepspeech-0.9.3-models/ node start.js
``` ```

2
nodejs_mic_vad_streaming/package.json
Просмотреть файл

@ -3,7 +3,7 @@
"version": "0.1.0", "version": "0.1.0",
"private": true, "private": true,
"dependencies": { "dependencies": {
"deepspeech": "^0.8.0", "deepspeech": "^0.9.3",
"mic": "^2.1.2", "mic": "^2.1.2",
"node-vad": "^1.1.4", "node-vad": "^1.1.4",
"speaker": "^0.5.1", "speaker": "^0.5.1",

2
nodejs_mic_vad_streaming/start.js
Просмотреть файл

@ -10,7 +10,7 @@ if (process.env.DEEPSPEECH_MODEL) {
DEEPSPEECH_MODEL = process.env.DEEPSPEECH_MODEL; DEEPSPEECH_MODEL = process.env.DEEPSPEECH_MODEL;
} }
else { else {
DEEPSPEECH_MODEL = __dirname + '/deepspeech-0.8.0-models'; DEEPSPEECH_MODEL = __dirname + '/deepspeech-0.9.3-models';
} }
let SILENCE_THRESHOLD = 200; // how many milliseconds of inactivity before processing the audio let SILENCE_THRESHOLD = 200; // how many milliseconds of inactivity before processing the audio

12
nodejs_wav/Readme.md
Просмотреть файл

@ -3,15 +3,15 @@
Download the pre-trained model (1.8GB): Download the pre-trained model (1.8GB):
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```
Edit references to models path if necessary: Edit references to models path if necessary:
``` ```
let modelPath = './models/deepspeech-0.8.0-models.pbmm'; let modelPath = './models/deepspeech-0.9.3-models.pbmm';
let scorerPath = './models/deepspeech-0.8.0-models.scorer'; let scorerPath = './models/deepspeech-0.9.3-models.scorer';
``` ```
Install Sox (for .wav file loading): Install Sox (for .wav file loading):
@ -23,8 +23,8 @@ brew install sox
Download test audio files: Download test audio files:
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/audio-0.8.0.tar.gz wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/audio-0.9.3.tar.gz
tar xfvz audio-0.8.0.tar.gz tar xfvz audio-0.9.3.tar.gz
``` ```
Install NPM dependencies: Install NPM dependencies:

4
nodejs_wav/index.js
Просмотреть файл

@ -5,13 +5,13 @@ const MemoryStream = require('memory-stream');
const Duplex = require('stream').Duplex; const Duplex = require('stream').Duplex;
const Wav = require('node-wav'); const Wav = require('node-wav');
let modelPath = './models/deepspeech-0.8.0-models.pbmm'; let modelPath = './models/deepspeech-0.9.3-models.pbmm';
let model = new DeepSpeech.Model(modelPath); let model = new DeepSpeech.Model(modelPath);
let desiredSampleRate = model.sampleRate(); let desiredSampleRate = model.sampleRate();
let scorerPath = './models/deepspeech-0.8.0-models.scorer'; let scorerPath = './models/deepspeech-0.9.3-models.scorer';
model.enableExternalScorer(scorerPath); model.enableExternalScorer(scorerPath);

2
nodejs_wav/package.json
Просмотреть файл

@ -8,7 +8,7 @@
}, },
"dependencies": { "dependencies": {
"argparse": "^1.0.10", "argparse": "^1.0.10",
"deepspeech": "0.8.0", "deepspeech": "0.9.3",
"node-wav": "0.0.2", "node-wav": "0.0.2",
"sox-stream": "^2.0.3", "sox-stream": "^2.0.3",
"util": "^0.11.1" "util": "^0.11.1"

8
uwp/DeepSpeechUWP/DeepSpeechUWP.csproj
Просмотреть файл

@ -159,20 +159,20 @@
</ItemGroup> </ItemGroup>
<ItemGroup> <ItemGroup>
<PackageReference Include="DeepSpeech"> <PackageReference Include="DeepSpeech">
<Version>0.8.0</Version> <Version>0.9.3</Version>
</PackageReference> </PackageReference>
<PackageReference Include="Microsoft.NETCore.UniversalWindowsPlatform"> <PackageReference Include="Microsoft.NETCore.UniversalWindowsPlatform">
<Version>6.2.9</Version> <Version>6.2.9</Version>
</PackageReference> </PackageReference>
<Reference Include="DeepSpeechClient, Version=1.0.0.0, Culture=neutral, processorArchitecture=AMD64"> <Reference Include="DeepSpeechClient, Version=1.0.0.0, Culture=neutral, processorArchitecture=AMD64">
<HintPath>..\packages\DeepSpeech.0.8.0\lib\net46\DeepSpeechClient.dll</HintPath> <HintPath>..\packages\DeepSpeech.0.9.3\lib\net46\DeepSpeechClient.dll</HintPath>
</Reference> </Reference>
</ItemGroup> </ItemGroup>
<ItemGroup> <ItemGroup>
<Content Include="models\deepspeech-0.8.0-models.pbmm"> <Content Include="models\deepspeech-0.9.3-models.pbmm">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content> </Content>
<Content Include="models\deepspeech-0.8.0-models.scorer"> <Content Include="models\deepspeech-0.9.3-models.scorer">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory> <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content> </Content>
</ItemGroup> </ItemGroup>

4
uwp/DeepSpeechUWP/MainPage.xaml.cs
Просмотреть файл

@ -65,8 +65,8 @@ namespace DeepSpeechUWP
{ {
string projectFolder = Directory.GetCurrentDirectory(); string projectFolder = Directory.GetCurrentDirectory();
string modelsFolder = Path.Combine(projectFolder, "models"); string modelsFolder = Path.Combine(projectFolder, "models");
string acousticModelPath = Path.Combine(modelsFolder, "deepspeech-0.8.0-models.pbmm"); string acousticModelPath = Path.Combine(modelsFolder, "deepspeech-0.9.3-models.pbmm");
string scorerPath = Path.Combine(modelsFolder, "deepspeech-0.8.0-models.scorer"); string scorerPath = Path.Combine(modelsFolder, "deepspeech-0.9.3-models.scorer");
client = new DeepSpeechClient.DeepSpeech(acousticModelPath); client = new DeepSpeechClient.DeepSpeech(acousticModelPath);
client.EnableExternalScorer(scorerPath); client.EnableExternalScorer(scorerPath);

2
vad_transcriber/requirements.txt
Просмотреть файл

@ -1,3 +1,3 @@
deepspeech==0.8.0 deepspeech==0.9.3
webrtcvad webrtcvad
pyqt5 pyqt5

4
web_microphone_websocket/Readme.md
Просмотреть файл

@ -6,8 +6,8 @@ to a NodeJS server and transmitting the DeepSpeech results back to the browser.
#### Download the pre-trained model (1.8GB): #### Download the pre-trained model (1.8GB):
``` ```
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.pbmm wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.pbmm
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.8.0/deepspeech-0.8.0-models.scorer wget https://github.com/mozilla/DeepSpeech/releases/download/v0.9.3/deepspeech-0.9.3-models.scorer
``` ```
#### Install: #### Install:

2
web_microphone_websocket/package.json
Просмотреть файл

@ -8,7 +8,7 @@
"@testing-library/user-event": "^7.1.2", "@testing-library/user-event": "^7.1.2",
"chai": "^4.2.0", "chai": "^4.2.0",
"chai-http": "^4.3.0", "chai-http": "^4.3.0",
"deepspeech": "^0.8.0", "deepspeech": "^0.9.3",
"defaults": "^1.0.3", "defaults": "^1.0.3",
"mocha": "^6.1.4", "mocha": "^6.1.4",
"node-vad": "^1.1.4", "node-vad": "^1.1.4",

2
web_microphone_websocket/server.js
Просмотреть файл

@ -3,7 +3,7 @@ const socketIO = require('socket.io');
const DeepSpeech = require('deepspeech'); const DeepSpeech = require('deepspeech');
const VAD = require('node-vad'); const VAD = require('node-vad');
let DEEPSPEECH_MODEL = __dirname + '/deepspeech-0.8.0-models'; // path to deepspeech english model directory let DEEPSPEECH_MODEL = __dirname + '/deepspeech-0.9.3-models'; // path to deepspeech english model directory
let SILENCE_THRESHOLD = 200; // how many milliseconds of inactivity before processing the audio let SILENCE_THRESHOLD = 200; // how many milliseconds of inactivity before processing the audio

2
web_microphone_websocket/test.sh
Просмотреть файл

@ -10,7 +10,7 @@ pushd ${THIS}
npm install $(get_npm_package_url) npm install $(get_npm_package_url)
npm install npm install
ln -s $HOME/DeepSpeech/models deepspeech-0.8.0-models ln -s $HOME/DeepSpeech/models deepspeech-0.9.3-models
yarn run test:client yarn run test:client
yarn run test:server yarn run test:server