SlideShare ist ein Scribd-Unternehmen logo
1 von 66
Downloaden Sie, um offline zu lesen
WIFI SSID:Spark+AISummit | Password: UnifiedDataAnalytics
Christian Grant
Listening at the Cocktail Party
with Deep Neural Networks and TensorFlow
#UnifiedDataAnalytics #SparkAISummit
tcpip001@gmail.com
Agenda
• The cocktail party problem
• Solving the cocktail party problem with deep neural networks (DNN’s)
• Future vision, and use cases
• Barriers to adopting deep neural networks at the edge
• Overcoming infrastructure, datasets, AI chips, and edge AI software
barriers
Real time speech separation at the edge
4#UnifiedDataAnalytics #SparkAISummit
Vision
Problem
5
When multiple people are speaking at the same time
• at a restaurant
• an airport
• or a cocktail party
Tuning in to one speaker is relatively easy for individuals with no hearing impairment.
Individuals with hearing impairment have difficulty in understanding speech in the
presence of competing voices.
6#UnifiedDataAnalytics #SparkAISummit
The Cocktail Party Problem
Problem - Mixed Audio
7
M1: It was a great Halloween party
F1: The nest was build with small twigs
Solution – Separated Tracks
8
M1: It was a great Halloween party F1: The nest was build with small twigs
Speech Separation Approach
9
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Speech Separation Approach
10
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Speech Separation Approach
11
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Speech Separation Approach
12
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Speech Separation Approach
13
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Speech Separation Approach
14
Input audio
Speaker 1 prediction
Speaker 1 source Speaker 2 source
Speaker 2 prediction
ISTFT
STFT – Short –time Fourier transform
ISTFT
Speaker 1 mask prediction
Deep Neural Network
Model
Weights
Tasks
15
Machine
Resource
Management
Process
Management
Data
Verification ML
Code
Configuration
Monitoring
Training
Transformations +
Feature Extraction
Inference
Partner
Collaboration
Publications
Evaluation
Data Collection
Tool & Platform Selection
Tasks
16
Platform
•Deep learning
virtual machine
•Real time
prediction
platform
•Demo platform
•Tiny platform
Data
•Generalized
data set
•Noise data
Transformation
•STFT
•ISTFT
•Spectrogram
Code
•Theano to
Keras + TF
•Keras + TF to
tf.keras
•Estimator API
•User friendly
code
Training
•HINT dataset
•Training lots of
models
Evaluation
•Lab listening
tests
•Metric
•Signal to
distortion ratio
Predictions
•Prediction
pipeline
•Predict 1000’s
of examples for
lots of models
Platform & Tool Selection
17
Tool Selection
18
Keras on Theano
No development
Keras on TensorFlow
•Keras
•Easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
TensorFlow Keras API
•Keras
•Very easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
•Distributed and local
•Keras models
•Google
•TensorFlow Lite
•TensorFlow Extended
•Production ready
Tool Selection
19
Keras on Theano
No development
Keras on TensorFlow
•Keras
•Easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
TensorFlow Keras API
•Keras
•Very easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
•Distributed and local
•Keras models
•Google
•TensorFlow Lite
•TensorFlow Extended
•Production ready
Tool Selection
20
Keras on Theano
No development
Keras on TensorFlow
•Keras
•Easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
TensorFlow Keras API
•Keras
•Very easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
•Distributed and local
•Keras models
•Google
•TensorFlow Lite
•TensorFlow Extended
•Production ready
Tool Selection
21
Keras on Theano
No development
Keras on TensorFlow
•Keras
•Easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
TensorFlow Keras API
•Keras
•Very easy to convert
•Google
•Large ecosystem
•TensorFlow Lite
•GPU
•Distributed and local
•Keras models
•Google
•TensorFlow Lite
•TensorFlow Extended
•Production ready
Data
22
Data
• 3 males + 3 females6 speakers
• 13 lists * 20 sentences * 6 speakers1560 sentences / files
• 2 – 3 seconds per sentence200 KB to 312 MB
• ~ 10 minutes of speech per speaker260 examples / speaker
• 16 bits per sample44.1 kHz sampling rate
Feature Engineering
24
Short-time Fourier Transform
# Load audio file
wav1, sr1 = librosa.load(‘voice.wav’, sr=None, mono=True, duration=2)
# Short-time Fourier transform
stft1 = librosa.stft(wav1)
Models
26
Fully Connected Neural Network
int_in = int(in_dim[0]) # 1032
inputs = Input(shape=(int_in,))
x = inputs
for i in range(n_hidden_layers): # 4 hidden layers
x = Dense(units=1024)(x)
x = Activation(‘sigmoid’)(x)
x = BatchNormalization()(x)
x = Dropout(dropout_val)(x)
int_out = int(op_dim[-1]) # 129
final_output = Dense(int_out, activation=‘sigmoid’)(x)
model = Model(inputs, final_output)
AO = tf.keras.optimizers.Adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=epsilon)
loss_func = 'mse'
model.compile(loss=loss_func, optimizer=AO)
Training
29
Training Configuration
• Topology: FDNN
• Speaker_1: M1
• Speaker_2: F1
• Epochs: 200
• Data_augmentation_ratio: 50
• Early_stopping: 25
• Data_input: HINT
• Training_list: L5 L6 L7 L8
• Validation_list: L9
• Testing_list: L1 L2
• Weights_output_directory:
Training
Data is generated batch-by-batch as tuples (inputs, outputs) by a Python generator
callbacks_list = [checkpoint, early_stopping, tensor_board]
history = model.fit_generator(generator,
steps_per_epoch=int(epoch_size / minbatchsize),
epochs=num_epochs,
validation_data=(Val_in, Val_out),
callbacks=callbacks_list,
verbose=2)
Evaluation and Prediction
32
Evaluation Configuration
• Train_topology: FDNN
• Train_speaker_1: M1
• Train_speaker_2: F1
• Pred_speaker_1: M1
• Pred_speaker_2: F1
• Data_directory: HINT
• Weights_directory:
• Pred_speech_type: Sentences
• Pred_list: L1 L2
• Pred_sentences: S1 S2
• Pred_output_directory:
Evaluation Results
34
Trained Pair
Predicted Pair M1/F1 M1/M2 F1/F2 M3/F3 ….
M1
F1
6.84
8.01
- - -
M1
M2
3.17
6.54
6.13
7.48
0.01
1.82
-
F1
F2
-0.38
0.72
-1.49
0.94
5.27
6.87
-
M3
F3
-3.94
11.23
-5.00
10.53
-8.50
7.56
0.33
13.07
FDNN
Signal to Distortion Ratio
Predicted Examples
35
Model Test example
(mixed audio)
Separated
tracks
SDR value Audio
result
M1F1_FDNN M1F1_L1L3_S2S8 M1 12.01
M1F1_FDNN M1F1_L1L3_S2S8 F1 14.21
M1F1_FDNN M3F3_L1L2_S12S20 M3 15.59
M1F1_FDNN M3F3_L1L2_S12S20 F3 -4.51
M1F1_FDNN M3F3_L1L3_S20S7 M3 -0.02
Future Direction
36
Real time speech separation at the edge
37#UnifiedDataAnalytics #SparkAISummit
Updated Vision
Use Cases
38
Related Use Cases
Speech to Speech
• Speech
separation
• Accessibility
• Noise removal
• Drive through
fast food /
cashier
Speech to Text
• Live transcription
• Air Traffic
Control
• Audio environment
classification
• Keyword speech
interfaces
• Speech
identification
• Speaker voice
identification
Speech to Text +
Text to Speech
• Dialogue systems
• Wake word
detection (keyword
/ trigger word
detection)
Devices and
Sensors
• AI pods, AI ear
buds
• AI headsets
• Hearables,
hearing aids
• Brain chips
• EEG, ECG
• Microphone arrays
Noise Removal
40#UnifiedDataAnalytics #SparkAISummit
Use Case
41
Live Transcription
42#UnifiedDataAnalytics #SparkAISummit
Use Case
43
Barriers
44
Barriers to Tiny ML Adoption
• Production environment
• Training
– Dataset
– Algorithm
• Inference
– Devices and chips
– Software
– Real time inference (latency < 20 milliseconds)
Production Environment
46
Production Tasks
Machine
Resource
Management
Process
Management
Data
Verification ML
Code
Data Collection
Configuration
Monitoring
Training
Transformations +
Feature Extraction
Inference
Evaluation
Tool & Platform Selection
Production Pipeline
Example
Validator
Trainer
Transform
StatisticsGenExampleGen
Model
Server
SchemaGen Evaluator
Model
Validator
Pusher
TF
Lite
Training &
Eval Data
TensorFlow Extended
MLFlow
Production Environment
Integrated Frontend for Job Management, Monitoring, Debugging, Data/Model/Evaluation Visualization
Configuration Framework and Job Orchestration
Tuner
Trainer
Data
Transformation
Data
analysis +
validation
Data
ingestion
Model
Evaluation Serving Logging
Pipeline Storage
Utilities for Garbage Collection, Data Access Controls
Serving
Analysis
AppStore
Production Reference Architecture
Data Lake
(Production Models
and Analytics)
Data
Labs
Sources
ERP
RDBMS
3RD
PARTY
Text
Sensors
Audio
and
Video
Machine
Logs
Web and
Social
IoT
PRIVATE HYBRIDPUBLIC
Services: Data Quality, Profiling, Retention, Reconciliation, Metadata, Security, Monitoring
Landing Cleansed Publishing/LOB
SIM EDW EDM
Ingestion
Staging Integration Calculation Semantic
Cloud Deployment
Analyst
Workbench
(Diagnostic)
DecisionMakers
Consumption
OperationalProcesses
Agile Data Lab Data Science Lab
Compute
Information
Portal
(Descriptive)
Calculation
Engine(s)
Data Lake
(Data Processing)
Advanced
Analytical
Engine(s)&Servers
Deep Learning Lab
EdgeDevices
DataScientist
(Predictive/
Prescriptive)
BusinessOperations
Integrated Frontend for Job Management, Monitoring, Debugging, Data/Model/Evaluation Visualization
Configuration Framework and Job Orchestration
Tuner
Trainer
Data
Transformation
Data
Analysis + Validation
Data
Ingestion
Model
Evaluation
Serving Logging
Pipeline Storage
Utilities for Garbage Collection, Data Access Controls
Serving
Analysis
Engineering Lab
Dataset
51
Improved Voice Dataset
HINT
• 1560 sentences
• 6 speakers
• 50% Male, 50%
Female
• 2 – 3 second
clips
• Studio recording
Build Your Own
Speech Dataset
• Download a few
hours of
speeches
• Chop the
speeches into 2
second clips ~ 1
million examples
• Preprocess as
needed
Common Voice
Mozilla
• Short sentences
• 30 GB
• 1087 hours
• 39,577 voices
• 23 % US English
• 9% UK English
• Open-source
voice database
Speech
Commands
Dataset
• 65000 utterances
• 30 short words
• 1 second clips
• Contributed by
the public through
the AIY website
AudioSet
Google
• 2,084,320 labeled
sound clips
• 10 second clips
• 1,01,065 speech
clips
• From YouTube
Improved Voice Dataset
HINT
• 1560 sentences
• 6 speakers
• 50% Male, 50%
Female
• 2 – 3 second
clips
• Studio recording
Build Your Own
Speech Dataset
• Download 1 hour
speeches from
multiple speakers
• Chop the
speeches into 2
second clips
• Preprocess as
needed
Common Voice
Mozilla
• Short sentences
• 30 GB
• 1087 hours
• 39,577 voices
• 23 % US English
• 9% UK English
• Open-source
voice database
Speech
Commands
Dataset
• 65000 utterances
• 30 short words
• 1 second clips
• Contributed by
the public through
the AIY website
AudioSet
Google
• 2,084,320 labeled
sound clips
• 10 second clips
• 1,01,065 speech
clips
• From YouTube
Improved Voice Datasets
HINT
• 1560 sentences
• 6 speakers
• 50% Male, 50%
Female
• 2 – 3 second
clips
• Studio recording
Build Your Own
Speech Dataset
• Download a few
hours of
speeches
• Chop the
speeches into 2
second clips ~ 1
million examples
• Preprocess as
needed
Common Voice
Mozilla
• Short sentences
• 30 GB
• 1087 hours
• 39,577 voices
• 23 % US English
• 9% UK English
• Open-source
voice database
Speech
Commands
Dataset
• 65000 utterances
• 30 short words
• 1 second clips
• Contributed by
the public through
the AIY website
AudioSet
Google
• 2,084,320 labeled
sound clips
• 10 second clips
• 1,01,065 speech
clips
• From YouTube
Approach
55
Improved Approach
Source: Looking to Listen: Audio-Visual Speech Separation (AI.GoogleBlog.com)
Tiny AI Devices & Chips
57
Tiny AI Devices
58
Cloud
•Training
•Inference
•DLVM
•GPU
Desktop
•Training (45m)
•Inference
•4 core I7
•GPU GTX 1080
Laptop
•Training (15h)
•Inference ok
•Laptop no GPU
•MacBook Pro
Smartphone
•Inference
•AI chips
•Coral Dev
Board
•Jetson Nano
Developer Kit
Edge
•Inference
•AI buds
•Hearables
AI Edge Chips
Google Edge TPU
• TPU
• TensorFlow Lite
• Image classification
• Object detection
• Mini-Go
• Deep feed forward
NN’s
• 4 TOPS
• 0.5 Watts / TOPS
Syntiant
• Neural decision
processor
• TensorFlow Lite
• 500000 parameters
• Wake word detection
• Speaker identification
• Keyword speech
interface
• Audio environment
classification
Smartphone NPU’s
• Snapdragon 855
• Kirin 990
59
Lightweight AI Software
60
Lightweight AI Software
• Model size
– Parameters
– Megabytes
• Optimizing the model
– Pruning
– Quantization
• Libraries
– TensorFlow Lite
– TensorFlow Lite (next version)
• Real Time (latency <= 20 milliseconds)
Summary
• The cocktail party problem
• Solving the cocktail party problem with deep neural networks (DNN’s)
• Future vision, and related use cases
• Barriers to adopting deep neural networks at the edge
• Overcoming infrastructure, datasets, AI chips, and edge AI software
barriers
Resources
• Simple Audio Recognition Tutorial
https://www.tensorflow.org/tutorials/sequences/audio_recognition
• Speaker and speech dependence in a deep neural networks speech separation algorithm, Eriksholm
https://wdh01.azureedge.net/-/media/eriksholm/main/files/publications/2019/bramslow-et-al-spin-2019-speaker-
and-speech-dependence-in-a-deep-neural-networks-speech-separation-a.pdf?la=en&rev=9978
• Why the Future of Machine Learning is Tiny, Pete Warden’s Blog
https://petewarden.com/2018/06/11/why-the-future-of-machine-learning-is-tiny/
• Looking to Listen: Audio-Visual Speech Separation
https://ai.googleblog.com/2018/04/looking-to-listen-audio-visual-speech.html
• Voice Datasets
– Common Voice: https://voice.mozilla.org/en
– AudioSet: https://research.google.com/audioset/
• Live Transcribe
– https://play.google.com/store/apps/details?id=com.google.audio.hearing.visualization.accessibility.scribe&hl=e
n_US
Q & A
64
Christian Grant
tcpip001@gmail.com
Thank You
#UnifiedDataAnalytics #SparkAISummit
DON’T FORGET TO RATE
AND REVIEW THE SESSIONS
SEARCH SPARK + AI SUMMIT

Weitere ähnliche Inhalte

Was ist angesagt?

Open nlp presentationss
Open nlp presentationssOpen nlp presentationss
Open nlp presentationssChandan Deb
 
Speaker recognition using MFCC
Speaker recognition using MFCCSpeaker recognition using MFCC
Speaker recognition using MFCCHira Shaukat
 
Microcontoller and Embedded System
Microcontoller and Embedded SystemMicrocontoller and Embedded System
Microcontoller and Embedded SystemKaran Thakkar
 
Deep learning for NLP and Transformer
 Deep learning for NLP  and Transformer Deep learning for NLP  and Transformer
Deep learning for NLP and TransformerArvind Devaraj
 
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...Universitat Politècnica de Catalunya
 
State space analysis shortcut rules, control systems,
State space analysis shortcut rules, control systems, State space analysis shortcut rules, control systems,
State space analysis shortcut rules, control systems, Prajakta Pardeshi
 
ARM Versions, architecture
ARM Versions, architectureARM Versions, architecture
ARM Versions, architectureKarthik Vivek
 
Microprocessor & microcontroller
Microprocessor & microcontroller Microprocessor & microcontroller
Microprocessor & microcontroller Nitesh Kumar
 
Lcd interfaing using 8051 and assambly language programming
Lcd interfaing using 8051 and assambly language programmingLcd interfaing using 8051 and assambly language programming
Lcd interfaing using 8051 and assambly language programmingVikas Dongre
 
Ec8352 signals and systems 2 marks with answers
Ec8352 signals and systems   2 marks with answersEc8352 signals and systems   2 marks with answers
Ec8352 signals and systems 2 marks with answersGayathri Krishnamoorthy
 
Design of Accumulator Unit
Design of Accumulator UnitDesign of Accumulator Unit
Design of Accumulator UnitHarshad Koshti
 
Sentiment Analysis on Twitter
Sentiment Analysis on TwitterSentiment Analysis on Twitter
Sentiment Analysis on TwitterSubarno Pal
 
Registradores contadores memórias_r2
Registradores contadores memórias_r2Registradores contadores memórias_r2
Registradores contadores memórias_r2Eduardo_borges
 
1909 BERT: why-and-how (CODE SEMINAR)
1909 BERT: why-and-how (CODE SEMINAR)1909 BERT: why-and-how (CODE SEMINAR)
1909 BERT: why-and-how (CODE SEMINAR)WarNik Chow
 
Lip Reading.pptx
Lip Reading.pptxLip Reading.pptx
Lip Reading.pptxNivethaT15
 
Sequential logic circuit optimization
Sequential logic circuit optimizationSequential logic circuit optimization
Sequential logic circuit optimizationAshis Kumar Chanda
 

Was ist angesagt? (20)

Open nlp presentationss
Open nlp presentationssOpen nlp presentationss
Open nlp presentationss
 
arithmetic ins in 8051
arithmetic ins in 8051arithmetic ins in 8051
arithmetic ins in 8051
 
Speaker recognition using MFCC
Speaker recognition using MFCCSpeaker recognition using MFCC
Speaker recognition using MFCC
 
Microcontoller and Embedded System
Microcontoller and Embedded SystemMicrocontoller and Embedded System
Microcontoller and Embedded System
 
Deep learning for NLP and Transformer
 Deep learning for NLP  and Transformer Deep learning for NLP  and Transformer
Deep learning for NLP and Transformer
 
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...
End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...
 
State space analysis shortcut rules, control systems,
State space analysis shortcut rules, control systems, State space analysis shortcut rules, control systems,
State space analysis shortcut rules, control systems,
 
ARM Versions, architecture
ARM Versions, architectureARM Versions, architecture
ARM Versions, architecture
 
Microprocessor & microcontroller
Microprocessor & microcontroller Microprocessor & microcontroller
Microprocessor & microcontroller
 
Lcd interfaing using 8051 and assambly language programming
Lcd interfaing using 8051 and assambly language programmingLcd interfaing using 8051 and assambly language programming
Lcd interfaing using 8051 and assambly language programming
 
Ec8352 signals and systems 2 marks with answers
Ec8352 signals and systems   2 marks with answersEc8352 signals and systems   2 marks with answers
Ec8352 signals and systems 2 marks with answers
 
Design of Accumulator Unit
Design of Accumulator UnitDesign of Accumulator Unit
Design of Accumulator Unit
 
Sentiment Analysis on Twitter
Sentiment Analysis on TwitterSentiment Analysis on Twitter
Sentiment Analysis on Twitter
 
Registradores contadores memórias_r2
Registradores contadores memórias_r2Registradores contadores memórias_r2
Registradores contadores memórias_r2
 
1909 BERT: why-and-how (CODE SEMINAR)
1909 BERT: why-and-how (CODE SEMINAR)1909 BERT: why-and-how (CODE SEMINAR)
1909 BERT: why-and-how (CODE SEMINAR)
 
Unit4.tms320c54x
Unit4.tms320c54xUnit4.tms320c54x
Unit4.tms320c54x
 
Speaker Recognition
Speaker RecognitionSpeaker Recognition
Speaker Recognition
 
Deeplearning in finance
Deeplearning in financeDeeplearning in finance
Deeplearning in finance
 
Lip Reading.pptx
Lip Reading.pptxLip Reading.pptx
Lip Reading.pptx
 
Sequential logic circuit optimization
Sequential logic circuit optimizationSequential logic circuit optimization
Sequential logic circuit optimization
 

Ähnlich wie Listening at the Cocktail Party with Deep Neural Networks and TensorFlow

Track 1 session 2 - st dev con 2016 - dsp concepts - innovating iot+wearab...
Track 1   session 2 - st dev con 2016 -  dsp concepts - innovating iot+wearab...Track 1   session 2 - st dev con 2016 -  dsp concepts - innovating iot+wearab...
Track 1 session 2 - st dev con 2016 - dsp concepts - innovating iot+wearab...ST_World
 
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...Edge AI and Vision Alliance
 
Deep Learning - Speaker Verification, Sound Event Detection
Deep Learning - Speaker Verification, Sound Event DetectionDeep Learning - Speaker Verification, Sound Event Detection
Deep Learning - Speaker Verification, Sound Event DetectionSai Kiran Kadam
 
Data Onboarding Breakout Session
Data Onboarding Breakout SessionData Onboarding Breakout Session
Data Onboarding Breakout SessionSplunk
 
Voice Recognition Accelerometers
Voice Recognition AccelerometersVoice Recognition Accelerometers
Voice Recognition AccelerometersNathan Glatz
 
Dante Audio Networking Fundamentals
Dante Audio Networking FundamentalsDante Audio Networking Fundamentals
Dante Audio Networking FundamentalsrAVe [PUBS]
 
The Big Data Is A Significant Subject Of Modern Times With...
The Big Data Is A Significant Subject Of Modern Times With...The Big Data Is A Significant Subject Of Modern Times With...
The Big Data Is A Significant Subject Of Modern Times With...Sarah Gordon
 
Audio Fingerprinting Introduction
Audio Fingerprinting IntroductionAudio Fingerprinting Introduction
Audio Fingerprinting IntroductionVikesh Khanna
 
OSINT RF Reverse Engineering by Marc Newlin
OSINT RF Reverse Engineering by Marc NewlinOSINT RF Reverse Engineering by Marc Newlin
OSINT RF Reverse Engineering by Marc NewlinEC-Council
 
Curriculum Development of an Audio Processing Laboratory Course
Curriculum Development of an Audio Processing Laboratory CourseCurriculum Development of an Audio Processing Laboratory Course
Curriculum Development of an Audio Processing Laboratory Coursesipij
 
DEFCON 23 - Ian Latter - remote access the apt
DEFCON 23 - Ian Latter - remote access the aptDEFCON 23 - Ian Latter - remote access the apt
DEFCON 23 - Ian Latter - remote access the aptFelipe Prado
 
Internet Of Things: Hands on: YOW! night
Internet Of Things: Hands on: YOW! nightInternet Of Things: Hands on: YOW! night
Internet Of Things: Hands on: YOW! nightAndy Gelme
 
Luiz eduardo. introduction to mobile snitch
Luiz eduardo. introduction to mobile snitchLuiz eduardo. introduction to mobile snitch
Luiz eduardo. introduction to mobile snitchYury Chemerkin
 
Polycom soundpoint ip601 data sheet
Polycom soundpoint ip601 data sheetPolycom soundpoint ip601 data sheet
Polycom soundpoint ip601 data sheetbest4systems
 
Advertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileAdvertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileDatabricks
 
[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic
[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic
[DSC Europe 22] Make some noise for AI in JavaScript - Sead DelalicDataScienceConferenc1
 
2018 FRSecure CISSP Mentor Program- Session 7
2018 FRSecure CISSP Mentor Program- Session 72018 FRSecure CISSP Mentor Program- Session 7
2018 FRSecure CISSP Mentor Program- Session 7FRSecure
 
Evolución de soluciones de colaboración | Diana Mate - VoIP2DAY 2017
Evolución de soluciones de colaboración |  Diana Mate - VoIP2DAY 2017Evolución de soluciones de colaboración |  Diana Mate - VoIP2DAY 2017
Evolución de soluciones de colaboración | Diana Mate - VoIP2DAY 2017VOIP2DAY
 

Ähnlich wie Listening at the Cocktail Party with Deep Neural Networks and TensorFlow (20)

Lect1_ DSP.pptx
Lect1_ DSP.pptxLect1_ DSP.pptx
Lect1_ DSP.pptx
 
Track 1 session 2 - st dev con 2016 - dsp concepts - innovating iot+wearab...
Track 1   session 2 - st dev con 2016 -  dsp concepts - innovating iot+wearab...Track 1   session 2 - st dev con 2016 -  dsp concepts - innovating iot+wearab...
Track 1 session 2 - st dev con 2016 - dsp concepts - innovating iot+wearab...
 
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...
“Comparing ML-Based Audio with ML-Based Vision: An Introduction to ML Audio f...
 
Deep Learning - Speaker Verification, Sound Event Detection
Deep Learning - Speaker Verification, Sound Event DetectionDeep Learning - Speaker Verification, Sound Event Detection
Deep Learning - Speaker Verification, Sound Event Detection
 
Data Onboarding Breakout Session
Data Onboarding Breakout SessionData Onboarding Breakout Session
Data Onboarding Breakout Session
 
Voice Recognition Accelerometers
Voice Recognition AccelerometersVoice Recognition Accelerometers
Voice Recognition Accelerometers
 
Dante Audio Networking Fundamentals
Dante Audio Networking FundamentalsDante Audio Networking Fundamentals
Dante Audio Networking Fundamentals
 
The Big Data Is A Significant Subject Of Modern Times With...
The Big Data Is A Significant Subject Of Modern Times With...The Big Data Is A Significant Subject Of Modern Times With...
The Big Data Is A Significant Subject Of Modern Times With...
 
Audio Fingerprinting Introduction
Audio Fingerprinting IntroductionAudio Fingerprinting Introduction
Audio Fingerprinting Introduction
 
OSINT RF Reverse Engineering by Marc Newlin
OSINT RF Reverse Engineering by Marc NewlinOSINT RF Reverse Engineering by Marc Newlin
OSINT RF Reverse Engineering by Marc Newlin
 
Curriculum Development of an Audio Processing Laboratory Course
Curriculum Development of an Audio Processing Laboratory CourseCurriculum Development of an Audio Processing Laboratory Course
Curriculum Development of an Audio Processing Laboratory Course
 
DEFCON 23 - Ian Latter - remote access the apt
DEFCON 23 - Ian Latter - remote access the aptDEFCON 23 - Ian Latter - remote access the apt
DEFCON 23 - Ian Latter - remote access the apt
 
Internet Of Things: Hands on: YOW! night
Internet Of Things: Hands on: YOW! nightInternet Of Things: Hands on: YOW! night
Internet Of Things: Hands on: YOW! night
 
Luiz eduardo. introduction to mobile snitch
Luiz eduardo. introduction to mobile snitchLuiz eduardo. introduction to mobile snitch
Luiz eduardo. introduction to mobile snitch
 
Polycom soundpoint ip601 data sheet
Polycom soundpoint ip601 data sheetPolycom soundpoint ip601 data sheet
Polycom soundpoint ip601 data sheet
 
Advertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileAdvertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-Mobile
 
09
0909
09
 
[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic
[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic
[DSC Europe 22] Make some noise for AI in JavaScript - Sead Delalic
 
2018 FRSecure CISSP Mentor Program- Session 7
2018 FRSecure CISSP Mentor Program- Session 72018 FRSecure CISSP Mentor Program- Session 7
2018 FRSecure CISSP Mentor Program- Session 7
 
Evolución de soluciones de colaboración | Diana Mate - VoIP2DAY 2017
Evolución de soluciones de colaboración |  Diana Mate - VoIP2DAY 2017Evolución de soluciones de colaboración |  Diana Mate - VoIP2DAY 2017
Evolución de soluciones de colaboración | Diana Mate - VoIP2DAY 2017
 

Mehr von Databricks

DW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDatabricks
 
Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Databricks
 
Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Databricks
 
Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Databricks
 
Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Databricks
 
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of HadoopDatabricks
 
Democratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDemocratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDatabricks
 
Learn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceLearn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceDatabricks
 
Why APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringWhy APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringDatabricks
 
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixThe Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixDatabricks
 
Stage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationStage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationDatabricks
 
Simplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchSimplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchDatabricks
 
Scaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesScaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesDatabricks
 
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesScaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesDatabricks
 
Sawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsSawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsDatabricks
 
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkRedis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkDatabricks
 
Re-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkRe-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkDatabricks
 
Raven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesRaven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesDatabricks
 
Processing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkProcessing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkDatabricks
 
Massive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeMassive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeDatabricks
 

Mehr von Databricks (20)

DW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptxDW Migration Webinar-March 2022.pptx
DW Migration Webinar-March 2022.pptx
 
Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1Data Lakehouse Symposium | Day 1 | Part 1
Data Lakehouse Symposium | Day 1 | Part 1
 
Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2Data Lakehouse Symposium | Day 1 | Part 2
Data Lakehouse Symposium | Day 1 | Part 2
 
Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2Data Lakehouse Symposium | Day 2
Data Lakehouse Symposium | Day 2
 
Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4Data Lakehouse Symposium | Day 4
Data Lakehouse Symposium | Day 4
 
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
5 Critical Steps to Clean Your Data Swamp When Migrating Off of Hadoop
 
Democratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized PlatformDemocratizing Data Quality Through a Centralized Platform
Democratizing Data Quality Through a Centralized Platform
 
Learn to Use Databricks for Data Science
Learn to Use Databricks for Data ScienceLearn to Use Databricks for Data Science
Learn to Use Databricks for Data Science
 
Why APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML MonitoringWhy APM Is Not the Same As ML Monitoring
Why APM Is Not the Same As ML Monitoring
 
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch FixThe Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
The Function, the Context, and the Data—Enabling ML Ops at Stitch Fix
 
Stage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI IntegrationStage Level Scheduling Improving Big Data and AI Integration
Stage Level Scheduling Improving Big Data and AI Integration
 
Simplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorchSimplify Data Conversion from Spark to TensorFlow and PyTorch
Simplify Data Conversion from Spark to TensorFlow and PyTorch
 
Scaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on KubernetesScaling your Data Pipelines with Apache Spark on Kubernetes
Scaling your Data Pipelines with Apache Spark on Kubernetes
 
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark PipelinesScaling and Unifying SciKit Learn and Apache Spark Pipelines
Scaling and Unifying SciKit Learn and Apache Spark Pipelines
 
Sawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature AggregationsSawtooth Windows for Feature Aggregations
Sawtooth Windows for Feature Aggregations
 
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen SinkRedis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
Redis + Apache Spark = Swiss Army Knife Meets Kitchen Sink
 
Re-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and SparkRe-imagine Data Monitoring with whylogs and Spark
Re-imagine Data Monitoring with whylogs and Spark
 
Raven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction QueriesRaven: End-to-end Optimization of ML Prediction Queries
Raven: End-to-end Optimization of ML Prediction Queries
 
Processing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache SparkProcessing Large Datasets for ADAS Applications using Apache Spark
Processing Large Datasets for ADAS Applications using Apache Spark
 
Massive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta LakeMassive Data Processing in Adobe Using Delta Lake
Massive Data Processing in Adobe Using Delta Lake
 

Kürzlich hochgeladen

Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...
Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...
Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...HyderabadDolls
 
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...gajnagarg
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...gajnagarg
 
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...HyderabadDolls
 
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxRESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxronsairoathenadugay
 
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...nirzagarg
 
Gartner's Data Analytics Maturity Model.pptx
Gartner's Data Analytics Maturity Model.pptxGartner's Data Analytics Maturity Model.pptx
Gartner's Data Analytics Maturity Model.pptxchadhar227
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNKTimothy Spann
 
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...gragchanchal546
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabiaahmedjiabur940
 
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...Health
 
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...HyderabadDolls
 
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...HyderabadDolls
 
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...gajnagarg
 
Top Call Girls in Balaghat 9332606886Call Girls Advance Cash On Delivery Ser...
Top Call Girls in Balaghat  9332606886Call Girls Advance Cash On Delivery Ser...Top Call Girls in Balaghat  9332606886Call Girls Advance Cash On Delivery Ser...
Top Call Girls in Balaghat 9332606886Call Girls Advance Cash On Delivery Ser...kumargunjan9515
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...nirzagarg
 
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...Valters Lauzums
 
Computer science Sql cheat sheet.pdf.pdf
Computer science Sql cheat sheet.pdf.pdfComputer science Sql cheat sheet.pdf.pdf
Computer science Sql cheat sheet.pdf.pdfSayantanBiswas37
 
Fun all Day Call Girls in Jaipur 9332606886 High Profile Call Girls You Ca...
Fun all Day Call Girls in Jaipur   9332606886  High Profile Call Girls You Ca...Fun all Day Call Girls in Jaipur   9332606886  High Profile Call Girls You Ca...
Fun all Day Call Girls in Jaipur 9332606886 High Profile Call Girls You Ca...kumargunjan9515
 
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...SOFTTECHHUB
 

Kürzlich hochgeladen (20)

Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...
Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...
Charbagh + Female Escorts Service in Lucknow | Starting ₹,5K To @25k with A/C...
 
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
 
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...
Sealdah % High Class Call Girls Kolkata - 450+ Call Girl Cash Payment 8005736...
 
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxRESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
 
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
 
Gartner's Data Analytics Maturity Model.pptx
Gartner's Data Analytics Maturity Model.pptxGartner's Data Analytics Maturity Model.pptx
Gartner's Data Analytics Maturity Model.pptx
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
 
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...
Gulbai Tekra * Cheap Call Girls In Ahmedabad Phone No 8005736733 Elite Escort...
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
 
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...
+97470301568>>weed for sale in qatar ,weed for sale in dubai,weed for sale in...
 
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
 
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...
Jodhpur Park | Call Girls in Kolkata Phone No 8005736733 Elite Escort Service...
 
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Indore [ 7014168258 ] Call Me For Genuine Models We...
 
Top Call Girls in Balaghat 9332606886Call Girls Advance Cash On Delivery Ser...
Top Call Girls in Balaghat  9332606886Call Girls Advance Cash On Delivery Ser...Top Call Girls in Balaghat  9332606886Call Girls Advance Cash On Delivery Ser...
Top Call Girls in Balaghat 9332606886Call Girls Advance Cash On Delivery Ser...
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
 
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...
Digital Advertising Lecture for Advanced Digital & Social Media Strategy at U...
 
Computer science Sql cheat sheet.pdf.pdf
Computer science Sql cheat sheet.pdf.pdfComputer science Sql cheat sheet.pdf.pdf
Computer science Sql cheat sheet.pdf.pdf
 
Fun all Day Call Girls in Jaipur 9332606886 High Profile Call Girls You Ca...
Fun all Day Call Girls in Jaipur   9332606886  High Profile Call Girls You Ca...Fun all Day Call Girls in Jaipur   9332606886  High Profile Call Girls You Ca...
Fun all Day Call Girls in Jaipur 9332606886 High Profile Call Girls You Ca...
 
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
 

Listening at the Cocktail Party with Deep Neural Networks and TensorFlow

  • 1. WIFI SSID:Spark+AISummit | Password: UnifiedDataAnalytics
  • 2. Christian Grant Listening at the Cocktail Party with Deep Neural Networks and TensorFlow #UnifiedDataAnalytics #SparkAISummit tcpip001@gmail.com
  • 3. Agenda • The cocktail party problem • Solving the cocktail party problem with deep neural networks (DNN’s) • Future vision, and use cases • Barriers to adopting deep neural networks at the edge • Overcoming infrastructure, datasets, AI chips, and edge AI software barriers
  • 4. Real time speech separation at the edge 4#UnifiedDataAnalytics #SparkAISummit Vision
  • 6. When multiple people are speaking at the same time • at a restaurant • an airport • or a cocktail party Tuning in to one speaker is relatively easy for individuals with no hearing impairment. Individuals with hearing impairment have difficulty in understanding speech in the presence of competing voices. 6#UnifiedDataAnalytics #SparkAISummit The Cocktail Party Problem
  • 7. Problem - Mixed Audio 7 M1: It was a great Halloween party F1: The nest was build with small twigs
  • 8. Solution – Separated Tracks 8 M1: It was a great Halloween party F1: The nest was build with small twigs
  • 9. Speech Separation Approach 9 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 10. Speech Separation Approach 10 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 11. Speech Separation Approach 11 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 12. Speech Separation Approach 12 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 13. Speech Separation Approach 13 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 14. Speech Separation Approach 14 Input audio Speaker 1 prediction Speaker 1 source Speaker 2 source Speaker 2 prediction ISTFT STFT – Short –time Fourier transform ISTFT Speaker 1 mask prediction Deep Neural Network Model Weights
  • 15. Tasks 15 Machine Resource Management Process Management Data Verification ML Code Configuration Monitoring Training Transformations + Feature Extraction Inference Partner Collaboration Publications Evaluation Data Collection Tool & Platform Selection
  • 16. Tasks 16 Platform •Deep learning virtual machine •Real time prediction platform •Demo platform •Tiny platform Data •Generalized data set •Noise data Transformation •STFT •ISTFT •Spectrogram Code •Theano to Keras + TF •Keras + TF to tf.keras •Estimator API •User friendly code Training •HINT dataset •Training lots of models Evaluation •Lab listening tests •Metric •Signal to distortion ratio Predictions •Prediction pipeline •Predict 1000’s of examples for lots of models
  • 17. Platform & Tool Selection 17
  • 18. Tool Selection 18 Keras on Theano No development Keras on TensorFlow •Keras •Easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU TensorFlow Keras API •Keras •Very easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU •Distributed and local •Keras models •Google •TensorFlow Lite •TensorFlow Extended •Production ready
  • 19. Tool Selection 19 Keras on Theano No development Keras on TensorFlow •Keras •Easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU TensorFlow Keras API •Keras •Very easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU •Distributed and local •Keras models •Google •TensorFlow Lite •TensorFlow Extended •Production ready
  • 20. Tool Selection 20 Keras on Theano No development Keras on TensorFlow •Keras •Easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU TensorFlow Keras API •Keras •Very easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU •Distributed and local •Keras models •Google •TensorFlow Lite •TensorFlow Extended •Production ready
  • 21. Tool Selection 21 Keras on Theano No development Keras on TensorFlow •Keras •Easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU TensorFlow Keras API •Keras •Very easy to convert •Google •Large ecosystem •TensorFlow Lite •GPU •Distributed and local •Keras models •Google •TensorFlow Lite •TensorFlow Extended •Production ready
  • 23. Data • 3 males + 3 females6 speakers • 13 lists * 20 sentences * 6 speakers1560 sentences / files • 2 – 3 seconds per sentence200 KB to 312 MB • ~ 10 minutes of speech per speaker260 examples / speaker • 16 bits per sample44.1 kHz sampling rate
  • 25. Short-time Fourier Transform # Load audio file wav1, sr1 = librosa.load(‘voice.wav’, sr=None, mono=True, duration=2) # Short-time Fourier transform stft1 = librosa.stft(wav1)
  • 27.
  • 28. Fully Connected Neural Network int_in = int(in_dim[0]) # 1032 inputs = Input(shape=(int_in,)) x = inputs for i in range(n_hidden_layers): # 4 hidden layers x = Dense(units=1024)(x) x = Activation(‘sigmoid’)(x) x = BatchNormalization()(x) x = Dropout(dropout_val)(x) int_out = int(op_dim[-1]) # 129 final_output = Dense(int_out, activation=‘sigmoid’)(x) model = Model(inputs, final_output) AO = tf.keras.optimizers.Adam(lr=lr, beta_1=beta_1, beta_2=beta_2, epsilon=epsilon) loss_func = 'mse' model.compile(loss=loss_func, optimizer=AO)
  • 30. Training Configuration • Topology: FDNN • Speaker_1: M1 • Speaker_2: F1 • Epochs: 200 • Data_augmentation_ratio: 50 • Early_stopping: 25 • Data_input: HINT • Training_list: L5 L6 L7 L8 • Validation_list: L9 • Testing_list: L1 L2 • Weights_output_directory:
  • 31. Training Data is generated batch-by-batch as tuples (inputs, outputs) by a Python generator callbacks_list = [checkpoint, early_stopping, tensor_board] history = model.fit_generator(generator, steps_per_epoch=int(epoch_size / minbatchsize), epochs=num_epochs, validation_data=(Val_in, Val_out), callbacks=callbacks_list, verbose=2)
  • 33. Evaluation Configuration • Train_topology: FDNN • Train_speaker_1: M1 • Train_speaker_2: F1 • Pred_speaker_1: M1 • Pred_speaker_2: F1 • Data_directory: HINT • Weights_directory: • Pred_speech_type: Sentences • Pred_list: L1 L2 • Pred_sentences: S1 S2 • Pred_output_directory:
  • 34. Evaluation Results 34 Trained Pair Predicted Pair M1/F1 M1/M2 F1/F2 M3/F3 …. M1 F1 6.84 8.01 - - - M1 M2 3.17 6.54 6.13 7.48 0.01 1.82 - F1 F2 -0.38 0.72 -1.49 0.94 5.27 6.87 - M3 F3 -3.94 11.23 -5.00 10.53 -8.50 7.56 0.33 13.07 FDNN Signal to Distortion Ratio
  • 35. Predicted Examples 35 Model Test example (mixed audio) Separated tracks SDR value Audio result M1F1_FDNN M1F1_L1L3_S2S8 M1 12.01 M1F1_FDNN M1F1_L1L3_S2S8 F1 14.21 M1F1_FDNN M3F3_L1L2_S12S20 M3 15.59 M1F1_FDNN M3F3_L1L2_S12S20 F3 -4.51 M1F1_FDNN M3F3_L1L3_S20S7 M3 -0.02
  • 37. Real time speech separation at the edge 37#UnifiedDataAnalytics #SparkAISummit Updated Vision
  • 39. Related Use Cases Speech to Speech • Speech separation • Accessibility • Noise removal • Drive through fast food / cashier Speech to Text • Live transcription • Air Traffic Control • Audio environment classification • Keyword speech interfaces • Speech identification • Speaker voice identification Speech to Text + Text to Speech • Dialogue systems • Wake word detection (keyword / trigger word detection) Devices and Sensors • AI pods, AI ear buds • AI headsets • Hearables, hearing aids • Brain chips • EEG, ECG • Microphone arrays
  • 41. 41
  • 43. 43
  • 45. Barriers to Tiny ML Adoption • Production environment • Training – Dataset – Algorithm • Inference – Devices and chips – Software – Real time inference (latency < 20 milliseconds)
  • 47. Production Tasks Machine Resource Management Process Management Data Verification ML Code Data Collection Configuration Monitoring Training Transformations + Feature Extraction Inference Evaluation Tool & Platform Selection
  • 49. Production Environment Integrated Frontend for Job Management, Monitoring, Debugging, Data/Model/Evaluation Visualization Configuration Framework and Job Orchestration Tuner Trainer Data Transformation Data analysis + validation Data ingestion Model Evaluation Serving Logging Pipeline Storage Utilities for Garbage Collection, Data Access Controls Serving Analysis
  • 50. AppStore Production Reference Architecture Data Lake (Production Models and Analytics) Data Labs Sources ERP RDBMS 3RD PARTY Text Sensors Audio and Video Machine Logs Web and Social IoT PRIVATE HYBRIDPUBLIC Services: Data Quality, Profiling, Retention, Reconciliation, Metadata, Security, Monitoring Landing Cleansed Publishing/LOB SIM EDW EDM Ingestion Staging Integration Calculation Semantic Cloud Deployment Analyst Workbench (Diagnostic) DecisionMakers Consumption OperationalProcesses Agile Data Lab Data Science Lab Compute Information Portal (Descriptive) Calculation Engine(s) Data Lake (Data Processing) Advanced Analytical Engine(s)&Servers Deep Learning Lab EdgeDevices DataScientist (Predictive/ Prescriptive) BusinessOperations Integrated Frontend for Job Management, Monitoring, Debugging, Data/Model/Evaluation Visualization Configuration Framework and Job Orchestration Tuner Trainer Data Transformation Data Analysis + Validation Data Ingestion Model Evaluation Serving Logging Pipeline Storage Utilities for Garbage Collection, Data Access Controls Serving Analysis Engineering Lab
  • 52. Improved Voice Dataset HINT • 1560 sentences • 6 speakers • 50% Male, 50% Female • 2 – 3 second clips • Studio recording Build Your Own Speech Dataset • Download a few hours of speeches • Chop the speeches into 2 second clips ~ 1 million examples • Preprocess as needed Common Voice Mozilla • Short sentences • 30 GB • 1087 hours • 39,577 voices • 23 % US English • 9% UK English • Open-source voice database Speech Commands Dataset • 65000 utterances • 30 short words • 1 second clips • Contributed by the public through the AIY website AudioSet Google • 2,084,320 labeled sound clips • 10 second clips • 1,01,065 speech clips • From YouTube
  • 53. Improved Voice Dataset HINT • 1560 sentences • 6 speakers • 50% Male, 50% Female • 2 – 3 second clips • Studio recording Build Your Own Speech Dataset • Download 1 hour speeches from multiple speakers • Chop the speeches into 2 second clips • Preprocess as needed Common Voice Mozilla • Short sentences • 30 GB • 1087 hours • 39,577 voices • 23 % US English • 9% UK English • Open-source voice database Speech Commands Dataset • 65000 utterances • 30 short words • 1 second clips • Contributed by the public through the AIY website AudioSet Google • 2,084,320 labeled sound clips • 10 second clips • 1,01,065 speech clips • From YouTube
  • 54. Improved Voice Datasets HINT • 1560 sentences • 6 speakers • 50% Male, 50% Female • 2 – 3 second clips • Studio recording Build Your Own Speech Dataset • Download a few hours of speeches • Chop the speeches into 2 second clips ~ 1 million examples • Preprocess as needed Common Voice Mozilla • Short sentences • 30 GB • 1087 hours • 39,577 voices • 23 % US English • 9% UK English • Open-source voice database Speech Commands Dataset • 65000 utterances • 30 short words • 1 second clips • Contributed by the public through the AIY website AudioSet Google • 2,084,320 labeled sound clips • 10 second clips • 1,01,065 speech clips • From YouTube
  • 56. Improved Approach Source: Looking to Listen: Audio-Visual Speech Separation (AI.GoogleBlog.com)
  • 57. Tiny AI Devices & Chips 57
  • 58. Tiny AI Devices 58 Cloud •Training •Inference •DLVM •GPU Desktop •Training (45m) •Inference •4 core I7 •GPU GTX 1080 Laptop •Training (15h) •Inference ok •Laptop no GPU •MacBook Pro Smartphone •Inference •AI chips •Coral Dev Board •Jetson Nano Developer Kit Edge •Inference •AI buds •Hearables
  • 59. AI Edge Chips Google Edge TPU • TPU • TensorFlow Lite • Image classification • Object detection • Mini-Go • Deep feed forward NN’s • 4 TOPS • 0.5 Watts / TOPS Syntiant • Neural decision processor • TensorFlow Lite • 500000 parameters • Wake word detection • Speaker identification • Keyword speech interface • Audio environment classification Smartphone NPU’s • Snapdragon 855 • Kirin 990 59
  • 61. Lightweight AI Software • Model size – Parameters – Megabytes • Optimizing the model – Pruning – Quantization • Libraries – TensorFlow Lite – TensorFlow Lite (next version) • Real Time (latency <= 20 milliseconds)
  • 62. Summary • The cocktail party problem • Solving the cocktail party problem with deep neural networks (DNN’s) • Future vision, and related use cases • Barriers to adopting deep neural networks at the edge • Overcoming infrastructure, datasets, AI chips, and edge AI software barriers
  • 63. Resources • Simple Audio Recognition Tutorial https://www.tensorflow.org/tutorials/sequences/audio_recognition • Speaker and speech dependence in a deep neural networks speech separation algorithm, Eriksholm https://wdh01.azureedge.net/-/media/eriksholm/main/files/publications/2019/bramslow-et-al-spin-2019-speaker- and-speech-dependence-in-a-deep-neural-networks-speech-separation-a.pdf?la=en&rev=9978 • Why the Future of Machine Learning is Tiny, Pete Warden’s Blog https://petewarden.com/2018/06/11/why-the-future-of-machine-learning-is-tiny/ • Looking to Listen: Audio-Visual Speech Separation https://ai.googleblog.com/2018/04/looking-to-listen-audio-visual-speech.html • Voice Datasets – Common Voice: https://voice.mozilla.org/en – AudioSet: https://research.google.com/audioset/ • Live Transcribe – https://play.google.com/store/apps/details?id=com.google.audio.hearing.visualization.accessibility.scribe&hl=e n_US
  • 66. DON’T FORGET TO RATE AND REVIEW THE SESSIONS SEARCH SPARK + AI SUMMIT