Voice Tech Podcast

Latest Episodes:

Your First Voice App - Hillary Long, Citro Digital Dec 09, 2018

Show notes

Hillary Long is the General Manager and Partner at CITRO Digital, a digital marketing agency, based in Allentown, PA, United States. Hillary takes us through the design and build process of their first voice app, the 'Naughty or Nice Quiz' for children. As an experienced technologist that's new to voice development, Hillary shares her experience from 'in the trenches', and the lessons her team learned.

We discuss content creation for a voice application, including the importance of iteration and beta testing, especially when working with children! Hillary shares her strategy for leveraging existing content held in a headless CMS, and how this was integrated with her voice application. We also cover the Alexa freetime program for Alexa kids devices, the Jovo cross platform framework for fast deployment to both Amazon Alexa and Google Assistant, and the significant differences between the release processes on both platforms.

Hillary also tells us about the creative ideas her team used to market their new voice app, including a competition for an Otterbox Google Assistant cover for kids, which you can enter if you're quick. It's a fun and relaxed interview that's packed with useful information for new and experienced voice developers alike. Have a listen and let me know what you think!

Links from the show:

Citro Digital: https://www.citrodigital.com
Naughty or Nice Quiz: http://bit.ly/citroquiz
Citro on Facebook: @citrodigital
Citro on Instagram: @Citro_Digital
Citro on Twitter: @Citro_Digital
Jovo: https://www.jovo.tech
Alexa FreeTime: http://amzn.to/alexafreetime
Contentful CMS: https://www.contentful.com/
Kentico CMS: https://www.kentico.com/
Digital Summit: https://philadelphia.digitalsummit.com/
Voice Summit: https://www.voicesummit.ai/

Find us here:

★ Support this podcast on Patreon ★

Conversational Marketing - Andy Mauro, Automat Nov 25, 2018

Show notes

Andy Mauro, CEO and co-founder of Automat, a company that helps brands deliver personalised one-to-one voice and messaging experiences to their customers. Andy has been helping to shape the voice industry for almost two decades now, having spent 16 years at Nuance, the most successful speech recognition company in history, and then TellMe, which was acquired by Microsoft.

In this episode we discuss Conversational Marketing and why it represents a fundamental shift in digital marketing. Andy explains how voice assistants will change the landscape of consumer sales & marketing, and the implications of voice search on brands' content marketing strategies. He founded Automat because the sales experience on web and mobile customer lack the personal touch, and he believes voice is the answer to this problem. We also discuss the role of influencers in conversational marketing, and how conversational marketing could impact brick and mortar shopping on the high street.

We also cover the history of the voice industry and how we got to where we are today, some of the lessons Andy has learned over his time in the industry, and why voice assistants might be heading for the trough of disillusionment in 2019. Thanks to Andy's wealth of experience it's an exceptionally interesting conversation, and highly relevant to anyone involved in the field.

Links from the show:

Automat: https://www.automat.ai/
Automat research report: https://www.automat.ai/opusresearch/
Automat Vichy case study: https://www.automat.ai/vichy-skinexpert/
Automat Cover Girl case study: https://www.automat.ai/covergirl/
Beauty Bot Directory: https://beauty.bot/directory/
Podcast w/conversational designer: http://bit.ly/2SaWimz
Automat article: http://bit.ly/2SanHoQ

Find us here:

★ Support this podcast on Patreon ★

Voice Emotion Analytics - Florian Eyben, audEERING Nov 11, 2018

Show notes

Florian Eyben is the CTO of audEERING, an audio analysis company that specialises in emotional artificial intelligence. Florian is a leading expert in voice emotion analytics, machine learning and signal processing, and our conversation is a great introduction to harnessing emotion in the voice. We discuss many of the incredible applications of voice emotion analytics, and dive into some projects carried out by audEERING for major brands in market research, call centers, social robotics, health and many more.

In the second half of the episode we explore openSMILE, audEERING's open source research toolkit for audio feature extraction. It is currently the state-of-the-art in affective computing for audio, and is the most widely-used tool for emotion recognition tasks in research and industry.

We also discuss the GeMAPS standard acoustic parameter recommendation, emotional dimensions such as valence and arousal, what data to use when developing emotion detection algorithms, and the future of multimodal emotion-enabled interfaces. It's an emotion-filled episode that you won't want to miss!

Links from the show:

Alexa skill building guide: https://voicetechpodcast.com/skillguide
audEERING site: https://www.audeering.com
audEERING on Facebook: http://bit.ly/aud-facebook
audEERING on Twitter: @audEERING
audEERING on Instagram: audEERING
audEERING on LinkedIn: http://bit.ly/aud-linkedin
openSMILE: https://opensmile.audeering.com
GeMAPS: https://ieeexplore.ieee.org/document/7160715
Florian Eyben Google Scholar: http://bit.ly/florian-eyben-scholar
ICASSP 2019: https://2019.ieeeicassp.org/
VoiceBot podcast: http://voicebot.ai

Find us here:

★ Support this podcast on Patreon ★

Open Source Voice Assistants - Joshua Montgomery, Mycroft Oct 28, 2018

Show notes

Joshua Montgomery is the CEO of Mycroft, the open source voice assistant. Mycroft is a much needed addition to the range of big tech voice assistants on the market, offering a new level of data privacy and customisation.

Their MkII device will be launched soon, and is available for pre-order now on IndieGogo. Josh describes how the MkII improves over previous iterations, takes us through the technical stack, and explains how you can start developing voice applications for the Mycroft ecosystem today.

We discuss the many benefits of open source for voice assistants, including scalability, censorship, data security and developer support. Josh reveals why many companies are unwilling to use Amazon Alexa, and why open source is the answer to this problem. This is a thought-provoking episode with a leader in the voice assistant space - enjoy!

Links from the show:

Full show notes: https://voicetechpodcast.com
Mycroft Homepage: https://mycroft.ai/
Get a Mycroft MkII on IndieGogo: https://igg.me/at/mycroft/x
Invest in Mycroft: https://www.startengine.com/mycroft-ai
Mycroft Github: https://github.com/MycroftAI/
Alexa Data Services: https://www.amazon.jobs/en/teams/alexa-data-services

Find us here:

★ Support this podcast on Patreon ★

Fast Scalable Voice IoT Apps - Syed Ahmed, PubNub Oct 14, 2018

Show notes

Syed Ahmed is a Developer Advocate at PubNub, a global Data Stream Network and real-time infrastructure-as-a-service company based in San Francisco, California. PubNub's products allow developers to build realtime web, mobile, voice and IoT applications. Low network latency is especially important for voice enabled applications that integrate with IoT devices, as any delay during a conversational interaction can affect usability and frustrate users.

In this episode, Syed explains how PubNub solves the latency problem by using a serverless architecture of PubNub Blocks, and walks us through a simple real-world example of a voice enabled doorbell. We learn about the publisher-subscriber pattern that underpins this technology, and why building voice apps with PubNub is quicker, easier and much more scalable than other methods.

A nice surprise during this conversation was that using PubNub Blocks not only improves the network performance of voice apps, but can be used to add advanced features such as contextual memory between phrases, and even between voice devices on different platforms, allowing developers to build more advanced interactions. It's a great episode, so check it out!

Links from the show:

PubNub Blocks: https://www.pubnub.com/docs/blocks-catalog
PubNub project example: https://hackernoon.com/voice-alexa-and-realtime-making-interactions-into-lasting-experiences-fa46aab9503b
More articles from Syed: https://medium.com/@nxsyed
Publish-subscribe pattern: https://en.wikipedia.org/wiki/Publish%E2%80%93subscribe_pattern
Syed's code examples on Github: https://github.com/nxsyed
Syed on Twitter: https://twitter.com/nxsyed
Syed on Insta: https://www.instagram.com/nxsyed/
Syed's personal site: https://www.nxsyed.me/

Find us here:

★ Support this podcast on Patreon ★

Audio Branding & Sound Design - Sebastian Hanfland, Hanfland & Friends Sep 30, 2018

Show notes

Sebastian Hanfland is the CEO of the audio branding consultancy, Hanfland and Friends. His team help companies define their their brand in audio form, in order to promote brand recognition, increase the perceived quality of products, and influence customers.

Sebastian explains what audio branding is, and how he selects the right voices and sounds for a brand. We explore many interesting examples of audio branding for products, customer service, workplaces and more, and hear a demo of their latest audio branding project for Humanzing Technologies. We also cover the typical challenges faced on audio branding projects, and what brands can do to prepare for the voice technology revolution.

This fantastic episode tells you all you need to know about sound design and audio branding. It is essential listening, not only for people working with voice technologies, but also for designers, marketers and content producers from every industry.

Links from the show:

Hanfland and Friends website (English): https://www.hanfland-and-friends.com
Hanfland and Friends website (German): https://www.hanfland-und-freunde.de
Hanfland and Friends Alexa Skill Stubbs: https://www.hanfland-und-freunde.de/blog/
Hanfland and Friends Twitter: @HanflandFreunde
Humanizing Technologies Sound Branding: https://www.humanizing.com/de/sound-2/
Audio Spotlight by Holosonics: https://www.holosonics.com/
Amazon Polly: https://aws.amazon.com/polly/

Find us here:

★ Support this podcast on Patreon ★

Podcasts of the Future - Bryan Colligan, AlphaVoice Sep 16, 2018

Show notes

Bryan Colligan is the co-founder of AlphaVoice, the easiest way to get your podcast and audio content onto Amazon Alexa and Google Assistant. In an especially fun conversation, Bryan shares his vision of how the podcasting and voice technology worlds are set to collide, and how the implications will be felt by platforms, content producers and consumers alike.

He reveals the 4 main business models of content producers, explains why voice is the tech wave that will dominate, and foretells the impending convergence of media. We then explore the AlphaVoice product, and examine his plans to build the ultimate Q&A voice interface using content transcription and indexing.

We also touch on how Google entering the podcasting space has influenced his product vision, and tackle the important subject of voice search, including the new paradigms of voice to voice, and voice to video search. Whether you're a publisher, producer or consumer of digital content, this is an unmissable episode!

Links from the show:

AlphaVoice: https://www.alphavoice.io/
AlphaVoice RadioLab demo: https://www.youtube.com/watch?v=XGozZKOw9r0
AlphaVoice YouTube channel: https://www.youtube.com/channel/UC4kzTZDYmtNY3ZzORAgITuw
Google speakable schema: http://www.ditii.com/google-rolls-out-new-schema-markup-speakable-dataset-for-news-contents/85138/
spaCy NLP: https://spacy.io/

Find us here:

★ Support this podcast on Patreon ★

Hum a Fingerprint, Extract a Melody - Dogac Basaran, CNRS Sep 02, 2018

Show notes

This is the second part of my conversation with Dogac Basaran, a post-doctoral researcher at CNRS, the French national scientific research centre. If you missed the first part, you might want to go back and listen to the previous episode on Signal Processing Basics for Audio.

Today, in part 2 of 2, we explore Dogac's research into audio fingerprinting, alignment, and melody extraction. By analysing the magnitude of frequency peaks and their relative spacing, Dogac shows us how it's possible to create audio fingerprints that can be used to detect and match audio recordings, even if they contain noise or are incomplete. These fingerprints have a variety of uses, including aligning multiple recordings of a single speaker/performance, and identifying a particular recording.

We also discuss query by humming, the state-of-the-art technique that takes an audio fingerprint of a person humming a melody, and matches it to a database of music recordings. Dogac also explains why learning how to build neural networks has become an essential skill in this field.

Links from the show:

Full show notes : http://bit.ly/voicetechpodcast
Dogac Basaran on Github: https://github.com/dogacbasaran
Dogac Basaran's websites: https://dbasaran.wp.imt.fr/ and http://dogacbasaran.com/
Signal Processing MOOC on Coursera: https://www.coursera.org/learn/dsp
MATLAB: https://matlab.mathworks.com/

Find us here:

★ Support this podcast on Patreon ★

Signal Processing Basics for Audio - Dogac Basaran, CNRS Aug 19, 2018

Show notes

Dogac Basaran is a post-doctoral researcher at CNRS, the French national scientific research centre. Today, in part 1 of 2, Dogac gives us a crash course in signal processing, where we learn what signal processing is and discover some of its many applications.

Leveraging his teaching experience, Dogac uses simple language and real-world examples to explain the fundamental signal processing concepts that are used in voice technology today. He defines frequency, period, and stability, and describes how sound cards use sampling and the Nyquist theorem to convert analogue signals into digital.

He then recommends some good educational resources and software packages, so you can learn more about signal processing and get started right away on your own programming projects.

Correction [21/08/2018]: The term 'stationarity' was cited in the episode, but this should have been 'stability'.

Links from the show:

Dogac Basaran on Github: https://github.com/dogacbasaran
Dogac Basaran's websites: https://dbasaran.wp.imt.fr/ and http://dogacbasaran.com/
Signal Processing MOOC on Coursera: https://www.coursera.org/learn/dsp
MATLAB: https://matlab.mathworks.com/
Python Scipy STFT package: https://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.stft.html
Humming databases: Jang, ThinkIT, IOSCAS and Task2
ISMIR association: http://www.ismir.net/
ISMIR 2018 conference: http://ismir2018.ircam.fr/
CNRS: http://www.cnrs.fr/
IRCAM: https://www.ircam.fr/

Find us here:

★ Support this podcast on Patreon ★

Perception of Smiles in the Voice - Pablo Arias, IRCAM Aug 05, 2018

Show notes

Pablo Arias is a final-year PhD student in perception and cognitive science at the audio research lab, IRCAM, in Paris. We discuss Pablo's work on how people perceive smiling and non-smiling voices.

First Pablo explains what cognitive science, neuroscience and perception are, and why research into these areas is so important. He then takes us through the aims, methods, and results of his latest research paper into smiling in the voice, and we discuss the academic and technological implications of his work.

Pablo shares his advice for budding PhD researchers, and the importance of having a mentor. Then we 'go deep' on how to learn how to listen, and outline the big unanswered questions in the field of perception right now.

Links from the show:

Pablo's latest paper on smiling in the voice: https://www.cell.com/current-biology/fulltext/S0960-9822(18)30752-8
Pablo's papers on Google Scholar: https://scholar.google.fr/citations?user=6jMFwJQAAAAJ&hl=en&oi=ao
Pablo on Twitter: @PabloAriasMusic
Pablo's email: arias@ircam.fr
Pablo's Black Macaw record label : https://www.facebook.com/blackmacawprod/
CREAM team: http://cream.ircam.fr
IRCAM: https://www.ircam.fr
[book] This is your brain on music, J, Levitin: https://amzn.to/2IYWAbb
[book] Music, Language, and the Brain, Aniruddh D. Patel: https://amzn.to/2KXleeh
Neatebox / Welcome app: https://www.neatebox.com/

Find us here:

★ Support this podcast on Patreon ★

Our TOPPODCAST Picks

Follow Us

Stay Connected

Related Podcasts

Links

Stay Connected