$30 off During Our Annual Pro Sale. View Details »
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
GDG DevFest 2015 Poland - How to grow your own ...
Search
Szymon Nowak
November 28, 2015
Programming
0
64
GDG DevFest 2015 Poland - How to grow your own Babel fish
Video of the demo is available at
https://youtu.be/Tv8ilBOKS2o
Szymon Nowak
November 28, 2015
Tweet
Share
More Decks by Szymon Nowak
See All by Szymon Nowak
meet.js Katowice - ES6 Promises 101
szimek
1
230
Serving WebP images via content negotiation
szimek
1
84
JSConf EU 2015 - How to grow your own Babel fish
szimek
0
330
Other Decks in Programming
See All in Programming
チームをチームにするEM
hitode909
0
350
複数人でのCLI/Infrastructure as Codeの暮らしを良くする
shmokmt
5
2.3k
S3 VectorsとStrands Agentsを利用したAgentic RAGシステムの構築
tosuri13
6
360
The Past, Present, and Future of Enterprise Java
ivargrimstad
0
210
AtCoder Conference 2025「LLM時代のAHC」
imjk
2
540
dotfiles 式年遷宮 令和最新版
masawada
1
800
モデル駆動設計をやってみようワークショップ開催報告(Modeling Forum2025) / model driven design workshop report
haru860
0
270
非同期処理の迷宮を抜ける: 初学者がつまづく構造的な原因
pd1xx
1
740
AIエージェントを活かすPM術 AI駆動開発の現場から
gyuta
0
440
Claude Codeの「Compacting Conversation」を体感50%減! CLAUDE.md + 8 Skills で挑むコンテキスト管理術
kmurahama
1
560
Canon EOS R50 V と R5 Mark II 購入でみえてきた最近のデジイチ VR180 事情、そして VR180 静止画に活路を見出すまで
karad
0
130
Navigation 3: 적응형 UI를 위한 앱 탐색
fornewid
1
380
Featured
See All Featured
The #1 spot is gone: here's how to win anyway
tamaranovitovic
1
860
Navigating Algorithm Shifts & AI Overviews - #SMXNext
aleyda
0
1k
Lightning Talk: Beautiful Slides for Beginners
inesmontani
PRO
1
400
The Web Performance Landscape in 2024 [PerfNow 2024]
tammyeverts
12
980
DBのスキルで生き残る技術 - AI時代におけるテーブル設計の勘所
soudai
PRO
60
37k
RailsConf & Balkan Ruby 2019: The Past, Present, and Future of Rails at GitHub
eileencodes
141
34k
Data-driven link building: lessons from a $708K investment (BrightonSEO talk)
szymonslowik
0
840
Building Better People: How to give real-time feedback that sticks.
wjessup
370
20k
Taking LLMs out of the black box: A practical guide to human-in-the-loop distillation
inesmontani
PRO
3
1.9k
Product Roadmaps are Hard
iamctodd
PRO
55
12k
Designing Powerful Visuals for Engaging Learning
tmiket
0
180
Design of three-dimensional binary manipulators for pick-and-place task avoiding obstacles (IECON2024)
konakalab
0
310
Transcript
None
None
WEBRTC
demo
None
None
None
WORKS IN A BROWSER
USES OPEN STANDARDS
TOOK ONE DAY TO WRITE
None
HOW DOES IT WORK?
1. SENDS AUDIO AND VIDEO USING WEBRTC
webrtc = new SimpleWebRTC({ localVideoEl: 'localVideo', remoteVideosEl: 'remoteVideos', autoRequestMedia: true
}) webrtc.on('readyToCall', () => { webrtc.joinRoom('your awesome room name') })
webrtc = new SimpleWebRTC({ localVideoEl: 'localVideo', remoteVideosEl: 'remoteVideos', autoRequestMedia: true
}) webrtc.on('readyToCall', () => { webrtc.joinRoom('your awesome room name') })
webrtc = new SimpleWebRTC({ localVideoEl: 'localVideo', remoteVideosEl: 'remoteVideos', autoRequestMedia: true
}) webrtc.on('readyToCall', () => { webrtc.joinRoom('your awesome room name') })
webrtc = new SimpleWebRTC({ localVideoEl: 'localVideo', remoteVideosEl: 'remoteVideos', autoRequestMedia: true
}) webrtc.on('readyToCall', () => { webrtc.joinRoom('your awesome room name') })
webrtc = new SimpleWebRTC({ localVideoEl: 'localVideo', remoteVideosEl: 'remoteVideos', autoRequestMedia: true
}) webrtc.on('readyToCall', () => { webrtc.joinRoom('your awesome room name') })
2. RECOGNIZES WHAT YOU SAY USING WEB SPEECH
recognition = new SpeechRecognition() recognition.on('result', (result) => { // do
something with the result }) recognition.start()
recognition = new SpeechRecognition() recognition.on('result', (result) => { // do
something with the result }) recognition.start()
recognition = new SpeechRecognition() recognition.on('result', (result) => { // do
something with the result }) recognition.start()
3. SENDS TRANSCRIPT TO GOOGLE TRANSLATE
4. SENDS TRANSLATION TO THE OTHER PERSON USING WEBRTC
webrtc.sendDirectlyToAll( 'channel label', 'payload label', payload )
5. READS IT OUT LOUD USING WEB SPEECH
msg = new SpeechSynthesisUtterance() msg.lang = 'en-GB' msg.text = 'German
jokes are the wurst!' speechSynthesis.speak(msg)
msg = new SpeechSynthesisUtterance() msg.lang = 'en-GB' msg.text = 'German
jokes are the wurst!' speechSynthesis.speak(msg)
None
WEB SPEECH API INTRO
ONE-SHOT CONTINUOUS
SERVER CLIENT
None
WEB SPEECH API RANT
1. STOPS IF YOU TALK FOR TOO LONG
2. STOPS IF YOU DON’T TALK FOR TOO LONG
3. DOESN’T WORK WITH WEBRTC MEDIA STREAMS
recognition = new SpeechRecognition()
recognition = new SpeechRecognition(audioStream)
4. WORKS ONLY IN CHROME* * DESKTOP CHROME (crbug.com/427567)
None
GOOGLE NOW BY GOOGLE
CORTANA BY MICROSOFT
SIRI BY APPLE
FIREFOX POWERED BY POCKET SPHINX
None
WE DON’T NEED TO WAIT FOR BROWSER VENDORS
recognition = new window.SpeechRecognition(audioStream)
recognition = new window.SpeechRecognition(audioStream)
recognition = new acme.SpeechRecognition(audioStream)
FACEBOOK M BY FACEBOOK
HOUND BY SOUNDHOUND
DUER BY BAIDU
DRAGON BY NUANCE
IBM WATSON BY IBM
POCKETSPHINX.JS BY SYLVAIN CHEVALIER
None
WEBRTC-TRANSLATE.HEROKUAPP.COM GITHUB.COM/SZIMEK/WEBRTC-TRANSLATE DEMO: CODE:
SZYMON NOWAK @SZIMEK COWBELL LABS FREELETICS we’re hiring!
THANK YOU