How do multilingual Digital Humans work?

Guildhawk Voice is an AI video offering that allows you to create dynamic video content using an AI face and voice, based just on written materials. Polished videos can be created in hours, with no need for actors, physical space, or equipment, and can be hyper-localised through use of specific accents, a diverse range of avatars, and translation.

Guildhawk Voice allows you to present information in a different, more engaging way. And, if you want to make your videos highly personalised to your company and brand, we can even create avatars based on specific individuals, complete with case-by-case recording of an individual’s voice.

Guildhawk Voice uses sophisticated text-to-speech technology and video synthesis to turn written material into content spoken by a synthetic face and voice, with lip movements to match.

This can be combined with quality translation to produce videos in multiple languages, accents and tones. See below for more information on the languages currently available.

Each video can have branded image or video backgrounds (including PPTs), and can even include background music, depending on what you need your content to achieve.

As mentioned above, there is some human intervention involved in all Text Perfect service levels – to one degree or another.

Our linguists and engineers are also involved in training the machine to review for client-specific rules.

Outside of these specific interventions, Text Perfect works independently to review the text, identify issues and flag them, and then implements suggested changes once these have been approved.

In order to train the machine in advance of commencing work, we need a minimum of 200,000 words of bilingual data,

i.e. existing approved translations, along with their corresponding source files. Source files should be less than 100 MB in size.

Note: In the absence of this advance bilingual data, Guildhawk Aided works on the basis of a set of sector - and language-specific terminology bases in the first instance. If you have chosen our Gold service level or above, we will then need to set preferences based on your feedback on the output, as we would anticipate a number of iterations will be needed to ensure the content is fit for purpose.

The preferences you indicate are saved to the translation memory to train the machine for the next iteration. This iterative process would need to take place several times – typically an average of 5 times – until we reach the level of output that is desired.

Costs for Guildhawk Voice start at £100, and short videos can be turned around in 1-2 working days.

Guildhawk Aided is currently compatible with all of the following file types and programs.

Note: Should content need to be pushed automatically from a particular portal or platform directly to Guildhawk Aided for translation, we can create an API to enable this automated process.

Guildhawk Voice currently supports 39 languages, as well as multiple accents and language variants,

e.g. Portuguese for Portugal and for Brazil:

Arabic	Greek	Portuguese
Bengali	Hebrew	Romanian
Bulgarian	Hindi	Russian
Cantonese	Hungarian	Slovak
Croatian	Indonesian	Slovenian
Czech	Italian	Spanish
Danish	Japanese	Swedish
Dutch	Korean	Tamil
English	Latvian	Telugu
Filipino	Malay	Thai
Finnish	Mandarin	Turkish
French	Norwegian	Ukrainian
German	Polish	Vietnamese

Guildhawk maintains UKAS-accredited ISO27001 certification, which is focused on information security and includes processes for protecting the integrity and confidentiality of data. Any content that you share with us as part of the Guildhawk Voice offering will be handled according to these secure information management practices.

Guildhawk Voice’s AI avatars use a combination of natural language processing techniques and a text-to-speech (TTS) engine to convert text into natural-sounding speech across multiple languages, accents and dialects.

The avatars’ facial expressions are created using a deep neural network, which ensures facial expressions are in sync with the synthesised speech generated by the TTS engine.

Anywhere, anytime – Every business has moments where a reaction to the market is required immediately. A digital avatar presents the information a business needs to convey in an engaging, professional manner, within minutes.

Trust – Presenting information using a custom avatar creates a truly personalised experience between a business and its customers. Digital avatars represent a business in a way that’s credible and professionally engaging.

Time and cost savings –  Personalised services can be created quickly at the fraction of the cost with avatars, minimising human workload by removing the need to hire actors, a design team and film crew, in turn reducing operating costs and accelerating throughput.

Engagement – Presenting information in a digestible format such as a video improves engagement by as much as 4x when compared to the same content in written form.

Enhanced corporate training programs – Avatars allow HR departments to produce corporate training programmes at a fraction of the cost, while allowing employees to learn anywhere and across a large number of languages. As of 2019, LinkedIn states that 59% of corporate training budgets are spent on online courses. Digital avatars significantly reduce the cost of creating these training programs.

Aging populations – As the global population is estimated to double by 2050, an increasing number of senior citizens are looking to the Internet for personal enrichment and education. Avatars make digital content more accessible and create a simplified, engaging and human-like user experience.

Global market penetration – Artificial intelligence avatars extend digital content to a wider population, as information is presented in a variety of languages, accents and dialects. Clear synthesised speech improves engagement with those that have literacy and learning difficulties.

Creating digital content with a multilingual avatar has many benefits over using a real actor. The most obvious is the ease with which you can create videos – if you’re leveraging existing written content, you don’t even have to produce a script!  This gives businesses greater flexibility to react in near real-time to events or communicate urgent information to their customers or employees.

Digital content presented by a multilingual avatar also reduces costs dramatically when compared to creating content with a real actor. Costs can quickly start to escalate when producing even a simple traditional video – you’ve got fees for an actor (or multiple actors, if your video is multilingual), studio, film crew, makeup artists, etc. Digital avatars incur none of these costs, meaning huge savings for the content creator. 

Artificial Intelligence also enables a single avatar to speak a wide range of languages, including using local accents and dialects, reducing the communication and cultural barriers businesses often face in local markets.