Presented By O’Reilly and Intel Nervana
Put AI to work
September 17-18, 2017: Training
September 18-20, 2017: Tutorials & Conference
San Francisco, CA

Applications of neural-based models for conversational speech

Yishay Carmiel (IntelligentWire | Spoken Labs)
2:35pm–3:15pm Wednesday, September 20, 2017
Implementing AI
Location: Grand Ballroom Level: Beginner
Secondary topics:  Deep learning
Average rating: ****.
(4.00, 1 rating)

Prerequisite Knowledge

  • A basic knowledge of deep learning and natural language processing

What you'll learn

  • Gain firsthand experience of the challenges of and solutions needed for analyzing conversational speech
  • Explore current bleeding-edge techniques
  • Understand considerations for building an appropriate solution to analyze human conversations


There is no doubt deep learning has made its mark on conversational AI, especially in natural language processing. Today almost every achievement in language understanding is based on neural networks. Yishay Carmiel explains why analyzing conversational speech is still a challenging proposition despite all the recent breakthroughs in natural language processing.

To take just one example, chats and emails make up just 35% of costumer-care interactions; the remaining 65% of the interactions are still voice based. Thus, analyzing conversational speech and finding proper actionable patterns on the conversation is one of the most important problems today in conversational AI.

Yishay explores the challenges and shares some potential solutions to tackle some of these problems, focusing on neural-based models.

Topics include:

  • Filling in the gaps between text-based processing and conversational speech processing
  • Neural speech embeddings for better speech representation models, specifically for user identification and separation during a conversation
  • Recent achievements of neural-based models for speech recognition
  • Advantages and recent breakthroughs of memory networks in analyzing conversational speech
  • Combining speech recognition and natural language processing for better analysis of human conversations
  • Future research and challenges
Photo of Yishay Carmiel

Yishay Carmiel

IntelligentWire | Spoken Labs

Yishay Carmiel is the founder of IntelligentWire, a company that develops and implements industry-leading deep learning and AI technologies for automatic speech recognition (ASR), natural language processing (NLP), and advanced voice data extraction, and the head of Spoken Labs, the strategic artificial intelligence and machine learning research arm of Spoken Communications. Yishay and his teams are currently working on bleeding-edge innovations that make the real-time customer experience a reality—at scale. Yishay has nearly 20 years’ experience as an algorithm scientist and technology leader building large-scale machine learning algorithms and serving as a deep learning expert.