Description
Description (EN): The objective of the project is to develop artificial inteligence (AI) based Estonian-Latvian machine translation toolkit by combination of Tilde’s state-of-the-art machine translation technologies and Avatar’s extensive data pool and translation expertise. Project Background The idea of Estonian-Latvian/Latvian-Estonian machine translation tool has arisen from a necessity to improve communication between close geographic and trading partners. The new product will be based on neural machine translation technologies - one of the most recognizable uses of Artificial Intelligence (AI). The project partners have both the state-of-the-art technologies and large data sets needed to develop AI-based machine translation for quality translation between these two complex languages. On 21 August 2018, the first joint meeting of the Latvian and Estonian Cabinets of Ministers discussed issues of common interest for both neighbouring countries. During this meeting Tilde was invited to present the idea of applying AI based translation technologies to creating an Estonian-Latvian translation service. Representatives from the both governments were supportive of the idea and expressed an interest in future development steps. This proposal for the development of the EstLat Translator is the natural outcome of this historic meeting. The direct Estonian-Latvian translation toolkit will eliminate the need to use a third language, usually English or Russian, which makes the translations unreliable in cross-border communication and services. Through cutting-edge neural machine translation technology it is now possible to “train” computers on large volumes of translation data for direct translation from Latvian to Estonian and Estonian to Latvian. As a component of developing the technology, the project will foster better understanding of both the Latvian and Estonian languages, will further advance the technologies necessary to ensure that Estonian and Latvian are examples how technology can help smaller European languages stay relevant in the digital age, and strengthen cross-border cooperation through stakeholder involvement. Cross Border nature of the project In the project the Latvian company Tilde and the Estonian company Avatar will cooperate to develop software tool that will make communication between languages much easier. As one of the largest Estonian translation companies skilled in using translation technologies Avatar has accumulated a significant amount of translation data. Tilde has the knowhow and experience in neural machine translation technologies to turn these translations into a service that will help both Estonian and Latvian businesses and public institutions communicate with their cross-border clients. The project will provide a synergy of competences for development of unique innovative product that not only will benefit the project partners but will also strengthen business, political and social ties between Estonia and Latvia. Objectives The objective of the project is to develop the EstLat Translator by building on the existing competences of Avatar and Tilde companies. The new toolkit will address the market need for an efficient cross-border communication solution and will create new cross-border business opportunities for both partners. Both public and private entities will benefit from an Estonian-Latvian-Estonian machine translation service. It will provide both project partners with a unique product to expand their businesses and seize new market opportunities. Impact The proposed high-quality machine translation service has the potential to be needed and used by every business and agency that communicates with their Estonian or Latvian partners. For example, it will be useful to scan the media and follow changes in laws, translate product and service descriptions. The project will serve a broader purpose to ensure digital relevance of Estonian and Latvian languages and strengthen usage of these small European languages in the digital age. It will showcase the benefits and opportunities of direct machine translation between smaller European languages. In the longer run access to information, different types of content and services in one’s native language by using AI driven language technology tools will strengthen social contacts and facilitate business relationships between Estonia and Latvia. Main Activities Project partners will establish and maintain a strong working relationship. The project will be implemented through three work packages (1) Analysis and specification will result in user requirement analysis, functional specification and the technical specification. (2) Development of the EstLat Translator - core activity of the project and will include preparing data, developing the neural machine translation systems, development of the EstLat Translator and piloting. (3) Marketing and sales planning - defining value proposition and pricing of the EstLat Translator , client segmentation and marketing communication and product sales planning.
Read more Achievements (EN): The objective of the project was to develop artificial intelligence (AI) based Estonian-Latvian machine translation toolkit EstLat Translator by a combination of Tilde’s (LV) state-of-the-art machine translation technologies and Avatar’s (EE) extensive data pool and translation expertise.
The new toolkit addresses the market need for an efficient cross-border communication solution and creates new cross-border business opportunities. Both public and private entities benefits from the Estonian-Latvian-Estonian machine translation service. It provides both project partners with a unique product to expand their businesses and seize new market opportunities.
Main activities were:
Analysis and specification that resulted in user requirement analysis, definition and prioritization of user stories, functional design and technical specifications.
Industry research for data modelling – during this activity, the partners worked on the required data development activities. Significant resources were invested in data criteria modeling for data categories to be used as a part of complex systems, identification of suitable data sources for parallel and monolingual corpora, development of automatic and unsupervised data harvesting and preprocessing workflows, human evaluation and quality assurance, and processing of the data for development of neural machine translation engines.
Neural Machine Translation (NMT) Training: this activity allowed to develop the Estonian-Latvian and Latvian-Estonian neural machine translation systems. The focus and the main objective of this activity was to develop machine translation systems by relying on linguistic resources collected and processed during the data acquisition activity. To address data sparsity, novel methods for handling rare and unknown phenomena (e.g. named entities and terminology) were studied and applied to hybridize the neural machine translation models. Furthermore, the machine translation models were deployed to enable dynamic (on-the-fly) adaptation of the models allowing to use domain-specific resources such as bilingual terminology and named entity dictionaries. Language-specific natural language processing methods were applied to improve and fine-tune the neural machine translation models (e.g. by using linguistic input features, improving rare named entities or term detection and processing quality, etc.).
Development and deployment of the EstLat Translator. The main focus of this activity was to develop and deploy the EstLat Translator and all related services – a one-stop productivity tool for companies and the public sector with any functionality that may be needed for effective automated translation (between Estonian and Latvian) and computer-assisted translation via a management system for post-editing and direct translation orders.
Marketing and sales planning. This activity allowed to prepare a value proposition for the EstLat Translator according to a set of functions and usage scenarios of each function. Identification of the target groups, development of the marketing strategy and channels for reaching out to these groups were an integral part of all activities. For the marketing purposes, the partners employed various methodologies and tools: digital brochures and presentations, a common visual identity, digital marketing campaigns (Google Ads, LinkedIn, Facebook and others), e-mail marketing, press releases and the official launching event with prominent speakers like Ieva Ilves (Advisor on Digital Policy to the President of Latvia) and others.
The outcome of the project is the EstLat Translator toolkit. The new toolkit addresses the market need for an efficient cross-border communication solution, and will create new cross-border business opportunities.
Main achievements of the project:
By combining the effort, experience and skills, successful and fruitful cooperation between Avatar and Tilde to develop a new service to offer more added value to the current customers and extend the services provided by both companies in Estonia and Latvia;
Joint development of a public AI-based Estonian-Latvian machine translation toolkit. The toolkit gives access to simple text translation and entire document translation that preserves the original formatting, and any user has a possibility to request professional translation services. This also includes a sophisticated platform for professional users where it is possible to process simple translations, use a professional online translation environment with integrated CAT (Computer-Assisted Translation) tools, build and manage translation memories and manage translation orders.
Successful marketing campaigns to reach out to a large audience and new customers in both partner countries.
Read more
Website: https://old.estlat.eu/
Expected Results (EN): The share of entrepreneurs carrying out joint product or service development
Expected Outputs (EN): The project will contribute to the programme result indicator ‘The share of entrepreneurs carrying out joint product and service development’, as the companies (Avatar and Tilde) will join forces to create the unique EstLat Translator. The cooperation will be rooted in a mutually beneficial approach where both entrepreneurs will benefit commercially and economically as well as contributing to the public good.
It will be not only cooperation based on common objective, but it will also be based on mutual trust and confidence through promoting trans-border linguistic connectivity and fostering closer economic ties. Furthermore, the project will boost cross boarder cultural ties and nurture people-to-people exchanges in the programme area as in the early project state we will set up a team building activities for all members involved in the project.
The joint team building activities for the teams will have specific goals to further cooperation. During the project, partners will carry out a series of joint events to create a strong project team and knowledge exchange. In order to break the ice and encourage sustainable cooperation during the project and after the project has ended, we will make one joint workshop to get to know each other and build strong network among project partners across the border. The team building will build skills like communication, planning and conflict resolution.
On a technical level, Avatar will contribute to data gathering which is a crucial component for the development of EstLat Translator, Tilde will share the expertise and knowledge on how to develop and customize the artificial intelligence driven machine translation systems.
In addition, the project team from both companies will consider participation in relevant field events to promote the neural MT based EstLat Translator and the shared development approach for the system development. Furthermore, a common marketing approach will be developed.
Read more
Thematic information
Specific Objective:
More jointly-developed products and services in the programme area
Thematic Objective:
(03) enhancing the competitiveness of SMEs by:
Investment Priority:
(03d) supporting the capacity of SMEs to grow in regional, national and international markets, and to engage in innovation processes
Partners (3)
Lead Partner:
SIA Tilde (GBER art. 20; 50%)
Department:
Machine translation business unit
Address:
Vienības gatve 75a, LV-1004 Riga, Latvia
Legal status:
private
Website:
http://www.tilde.com
Total budget:
EUR 189 971.91
ERDF budget:
EUR 94 985.96
-
SIA Tilde (GBER art. 25; 75%)
-
Name:
SIA Tilde (GBER art. 25; 75%)
Department:
Machine translation business unit
Address:
Vienības 75a, LV-1004 Riga, Latvia
Legal status:
private
Website:
http://tilde.com
Total budget:
EUR 117 806.67
ERDF budget:
EUR 88 355.00
-
Avatar, OU (De Minimis, 85%)
-
Name:
Avatar, OU (De Minimis, 85%)
Address:
Tila küla, Keldri, 60532 Tartumaa, Tartu vald, Estonia
Legal status:
private
Website:
http://www.avatar.ee
Total budget:
EUR 214 211.07
ERDF budget:
EUR 182 079.41
