May 30, 2023
At Voice Tech Paris, ReadSpeaker presents the new uses of synthetic voices: custom brand voices, human emotion and multilingual interactive voice ads.
Jean-Rémi Larcelet-Prost4 min readParis, 9 November. At Voice Tech Paris, held on 23 and 24 November, ReadSpeaker will present the new uses of synthetic voices that conversational AI is bringing ever closer to human speech. Custom voices, the expression of human emotion and multilingual capability are opening the way to innovative and creative applications of text to speech.
Smart speakers, voice assistants and other connected objects with intelligence built in are driving the rise of the Internet of Voice. More than one French person in two uses a voice assistant, and 77 percent of them consider that it improves their daily life, according to the study run by La Poste with Isoskèle (opens in a new tab). Text to speech, increasingly present in the interactions between brands and their end users, is becoming a strategic way to stand apart from the competition.
Long robotic and unengaging, synthetic voices are becoming human. By combining text-to-speech technology with deep neural networks (DNN), ReadSpeaker creates synthetic voices that are hard to tell apart from human ones. Brands are then able to offer conversational experiences that are more natural and more appealing. Custom synthetic voices can express emotion, such as enthusiasm, joy, sadness or regret, and they can also carry a tone that belongs to a given field: media, gaming, education and others.
Bol.com, the Belgian and Dutch online retailer, recently launched its custom brand voice for Google Assistant. The voice identity created by ReadSpeaker sets it apart from the standard voices used by default on the Google platform. For any sales or customer service request, users simply say “Talk to bol.com” to reach the bol.com assistant. It answers questions, offers daily deals and keeps customers informed about the status of their earlier requests.
Another example is Spotify, which wanted to create a smooth multimodal experience for the users of its limited edition Car Thing (opens in a new tab), a smart player for the car. By giving its voice assistant a synthetic voice with a natural timbre, Spotify gives its brand identity a personality in sound.
“Voice is a component of brand identity, just like the name or the logo. It is both a landmark for the consumer and a point of difference. In the same way, the custom synthetic voice makes it possible to widen the customer base and to strengthen brand identity.”
Roy Lindemann, CMO of ReadSpeaker
Interactive voice ads are a form of brand engagement designed for digital audio channels: smart speakers, mobile streaming services and in-car infotainment systems. These ads speak directly to listeners and invite them to accept or decline an offer by answering with a single word. Because they ask listeners whether they want to know more, they are less intrusive and more effective. Their click-through rate can reach 7 to 10 percent, against 1 to 3 percent for a digital display ad.
The multilingual capability of ReadSpeaker synthetic voices plays a decisive part here, because it allows an interactive voice ad to be duplicated quickly and at lower cost. ReadSpeaker lets a brand be represented by a single voice that sounds native across a whole range of languages.
Attend the ReadSpeaker virtual workshops by registering online:
Voice Tech Paris (opens in a new tab) brings the B2B voice ecosystem together in a fully digital format, to inform and connect the professionals of the sector. At this large-scale event dedicated to enterprise applications of voice technology, organisations can discover the latest use cases and take the measure of what these technologies will mean for them.
ReadSpeaker, a leading provider of text-to-speech solutions, has created more than 110 different synthetic voices in 35 languages. It has built more than 10,000 digital voice interfaces for brands established in over 60 countries. Its text-to-speech technology, based on deep neural networks (DNN), delivers the most realistic and accurate custom synthetic voices on the market. With more than 20 years of experience, ReadSpeaker offers brands sophisticated text-to-speech solutions and specialist support for the creation of engaging voice interfaces.
Follow the company on Twitter (opens in a new tab) and LinkedIn (opens in a new tab).
Voice & AI
Cloud text-to-speech APIs are quick to prototype with, but regulated and latency-sensitive products often need more control.
Voice & AI
A custom voice is more than a synthetic voice model. For an enterprise, it can become a long-term communication asset used across products, customer services, applications and operational systems. This guide explains how enterprise custom voices are planned, created, governed and deployed.
Voice & AI
Why a standardized TTS strategy is the key to reducing costs and improving the user experience across a complex transportation network.
