5 AI tools for translation - Cointelegraph - Translation

Explore AI translation tools, their features, benefits and pricing models to find the right solution for your translation needs.

Overview

Join us on social networks

Translation is the process of converting written or spoken content from one language to another while preserving its meaning. By automating and enhancing the translation process, artificial intelligence (AI) has significantly contributed to changing the translation industry.

To evaluate and comprehend the structure, syntax and context of the source language and produce correct translations in the target language, AI-powered translation systems use machine learning algorithms and natural language processing techniques. 

Types of AI-powered translation systems

AI-powered translation systems can be categorized into two main approaches:

Rule-based machine translation (RBMT)

To translate text, RBMT systems use dictionaries and pre-established linguistic rules. Linguists and other experts create these guidelines and dictionaries that specify how to translate words, phrases and grammatical structures.

While RBMT systems are capable of producing accurate translations for some language pairs, they frequently face limitations due to the complexity and diversity of linguistic systems, which makes them less useful for translations that are more complex.

Statistical machine translation (SMT)

SMT systems employ statistical models that have been developed using sizable bilingual corpora. These algorithms analyze the words and phrases in the source and target languages to find patterns and correlations.

SMT systems are able to make educated assumptions about the ideal translation for a particular input by examining enormous volumes of data. With more training data, SMT systems get more accurate, although they may have trouble with unusual or rare phrases.

Neural machine translation (NMT) has recently become more well-known in the translation industry. To produce translations, NMT systems use deep learning methods, notably neural networks. Compared to earlier methods, these models are better able to represent the context, semantics and complexities of languages. NMT systems have proven to perform better than other technologies, and they are widely employed in many well-known translation services and applications.

Advantages of AI in translation

The use of AI in translation offers several advantages:

• Speed and efficiency: AI-powered translation systems can process large volumes of text quickly, accelerating the translation process and improving productivity.
• Consistency: AI ensures consistent translations by adhering to predefined rules and learned patterns, reducing errors and discrepancies.
• Customization and adaptability: AI models can be fine-tuned and customized for specific domains, terminologies or writing styles, resulting in more accurate and contextually appropriate translations.
• Continuous improvement: AI systems can learn from user feedback and update their translation models over time, gradually improving translation quality.

AI tools for translation

There are several AI tools available for translation that leverage machine learning and natural language processing techniques. Here are five popular AI tools for translation:

Google Translate

Google Translate is a widely used AI-powered translation tool. To offer translations for different language pairs, it combines rule-based and neural machine translation models. It offers functionalities for text translation, website translation and even speech-to-text and text-to-speech.

Google Translate offers both free and paid versions. The basic translation services, including text translation, website translation and basic speech-to-text features, are accessible to users for free. However, Google also offers a paid service called Google Translate API for developers and businesses with more extensive translation needs. API usage is subject to pricing based on the number of characters translated.

Microsoft Translator

Another capable AI translation tool is Microsoft Translator. It offers translation services for many different languages and makes use of neural machine translation models. It offers developers APIs and SDKs so they may incorporate translation functionality into their projects.

Microsoft Translator offers a tiered pricing model. It has a free tier that allows users to access basic translation services with certain limitations. Microsoft also provides paid plans for higher volume and advanced features. The pricing is typically based on the number of characters translated or the number of API requests made.

DeepL

DeepL is an AI-driven translation tool known for its high-quality translations. It utilizes neural machine translation models and claims to outperform other popular translation tools in terms of accuracy. DeepL supports multiple language pairs and offers a user-friendly interface.

DeepL offers both free and paid versions. The free version of DeepL allows users to access its translation services with certain usage restrictions. DeepL also offers a subscription-based premium plan called DeepL Pro, which provides additional benefits, such as faster translation speeds, unlimited usage and the ability to integrate the service into other applications.

Systran

Systran is a language technology company that provides AI-powered translation solutions. It offers a range of products and services, including neural machine translation engines, translation APIs and specialized industry solutions. Systran focuses on customization and domain-specific translations.

Pricing for Systran’s offerings is typically based on the specific requirements and level of customization desired by the client.

Trados Enterprise

RWS is a global leader in translation and localization services, and it provides various language technology solutions to support translation and multilingual content management. 

One of its language technology offerings is Trados Enterprise (previously RWS Language Cloud). This cloud-based platform is designed to streamline the translation process, enhance collaboration and improve translation quality. It provides a range of features and tools to manage translation projects, such as translation memory, terminology management, project management and linguistic assets.

Trados Enterprise offers different versions tailored to specific needs. The Studio version is priced at $125 per month and provides an industry-leading computer-assisted translation (CAT) tool for professional linguists. The Team version, priced at $185 per user per month, focuses on cloud-based collaboration for translation projects.

The Accelerate version starts at $365 per user per month and offers end-to-end translation management for organizations with custom requirements. RWS also provides a free trial for interested users and encourages potential customers to request a demo to explore their offerings in detail.

Adblock test (Why?)

5 AI tools for translation - Cointelegraph - Translation

Explore AI translation tools, their features, benefits and pricing models to find the right solution for your translation needs.

Overview

Join us on social networks

Translation is the process of converting written or spoken content from one language to another while preserving its meaning. By automating and enhancing the translation process, artificial intelligence (AI) has significantly contributed to changing the translation industry.

To evaluate and comprehend the structure, syntax and context of the source language and produce correct translations in the target language, AI-powered translation systems use machine learning algorithms and natural language processing techniques. 

Types of AI-powered translation systems

AI-powered translation systems can be categorized into two main approaches:

Rule-based machine translation (RBMT)

To translate text, RBMT systems use dictionaries and pre-established linguistic rules. Linguists and other experts create these guidelines and dictionaries that specify how to translate words, phrases and grammatical structures.

While RBMT systems are capable of producing accurate translations for some language pairs, they frequently face limitations due to the complexity and diversity of linguistic systems, which makes them less useful for translations that are more complex.

Statistical machine translation (SMT)

SMT systems employ statistical models that have been developed using sizable bilingual corpora. These algorithms analyze the words and phrases in the source and target languages to find patterns and correlations.

SMT systems are able to make educated assumptions about the ideal translation for a particular input by examining enormous volumes of data. With more training data, SMT systems get more accurate, although they may have trouble with unusual or rare phrases.

Neural machine translation (NMT) has recently become more well-known in the translation industry. To produce translations, NMT systems use deep learning methods, notably neural networks. Compared to earlier methods, these models are better able to represent the context, semantics and complexities of languages. NMT systems have proven to perform better than other technologies, and they are widely employed in many well-known translation services and applications.

Advantages of AI in translation

The use of AI in translation offers several advantages:

• Speed and efficiency: AI-powered translation systems can process large volumes of text quickly, accelerating the translation process and improving productivity.
• Consistency: AI ensures consistent translations by adhering to predefined rules and learned patterns, reducing errors and discrepancies.
• Customization and adaptability: AI models can be fine-tuned and customized for specific domains, terminologies or writing styles, resulting in more accurate and contextually appropriate translations.
• Continuous improvement: AI systems can learn from user feedback and update their translation models over time, gradually improving translation quality.

AI tools for translation

There are several AI tools available for translation that leverage machine learning and natural language processing techniques. Here are five popular AI tools for translation:

Google Translate

Google Translate is a widely used AI-powered translation tool. To offer translations for different language pairs, it combines rule-based and neural machine translation models. It offers functionalities for text translation, website translation and even speech-to-text and text-to-speech.

Google Translate offers both free and paid versions. The basic translation services, including text translation, website translation and basic speech-to-text features, are accessible to users for free. However, Google also offers a paid service called Google Translate API for developers and businesses with more extensive translation needs. API usage is subject to pricing based on the number of characters translated.

Microsoft Translator

Another capable AI translation tool is Microsoft Translator. It offers translation services for many different languages and makes use of neural machine translation models. It offers developers APIs and SDKs so they may incorporate translation functionality into their projects.

Microsoft Translator offers a tiered pricing model. It has a free tier that allows users to access basic translation services with certain limitations. Microsoft also provides paid plans for higher volume and advanced features. The pricing is typically based on the number of characters translated or the number of API requests made.

DeepL

DeepL is an AI-driven translation tool known for its high-quality translations. It utilizes neural machine translation models and claims to outperform other popular translation tools in terms of accuracy. DeepL supports multiple language pairs and offers a user-friendly interface.

DeepL offers both free and paid versions. The free version of DeepL allows users to access its translation services with certain usage restrictions. DeepL also offers a subscription-based premium plan called DeepL Pro, which provides additional benefits, such as faster translation speeds, unlimited usage and the ability to integrate the service into other applications.

Systran

Systran is a language technology company that provides AI-powered translation solutions. It offers a range of products and services, including neural machine translation engines, translation APIs and specialized industry solutions. Systran focuses on customization and domain-specific translations.

Pricing for Systran’s offerings is typically based on the specific requirements and level of customization desired by the client.

Trados Enterprise

RWS is a global leader in translation and localization services, and it provides various language technology solutions to support translation and multilingual content management. 

One of its language technology offerings is Trados Enterprise (previously RWS Language Cloud). This cloud-based platform is designed to streamline the translation process, enhance collaboration and improve translation quality. It provides a range of features and tools to manage translation projects, such as translation memory, terminology management, project management and linguistic assets.

Trados Enterprise offers different versions tailored to specific needs. The Studio version is priced at $125 per month and provides an industry-leading computer-assisted translation (CAT) tool for professional linguists. The Team version, priced at $185 per user per month, focuses on cloud-based collaboration for translation projects.

The Accelerate version starts at $365 per user per month and offers end-to-end translation management for organizations with custom requirements. RWS also provides a free trial for interested users and encourages potential customers to request a demo to explore their offerings in detail.

Adblock test (Why?)

Archaeologists use artificial intelligence (AI) to translate 5000-year-old cuneiform tablets - HeritageDaily - Translation

A team of archaeologists and computer scientists have created an AI program that can translate ancient cuneiform tablets instantly using neural machine learning translations.

In a paper published in the journal PNAS Nexus, from the Oxford University Press, the researchers have applied the AI program to translate Akkadian texts with a high level of accuracy.

Akkadian is an ancient East Semitic language, was once spoken in various regions of ancient Mesopotamia, including Akkad, Assyria, Isin, Larsa, Babylonia, and possibly Dilmun.

 

The language is preserved on clay tablets dating back to 2500 BC that was written using cuneiform, a script adopted from the Sumerians using wedge-shaped symbols pressed in wet clay.

According to the researchers: “Hundreds of thousands of clay tablets inscribed in the cuneiform script document the political, social, economic, and scientific history of ancient Mesopotamia. Yet, most of these documents remain untranslated and inaccessible due to their sheer number and limited quantity of experts able to read them.”

The AI program has a high level accuracy when translating formal Akkadian texts such as royal decrees or omens that follow a certain pattern. More literary and poetic texts, such as letters from priests or tracts, were more likely to have “hallucinations” – an AI term meaning that the machine generated a result completely unrelated to the text provided.

The goal of the neural machine translation (NMT) into English from Akkadian is to be part of a human–machine collaboration, by creating a pipeline that assists the scholar or student of the ancient language.

 

Currently, the NMT model is available on an online notebook and the source code has been made available on GitHub at Akkademia. The researchers are currently developing an online application called the Babylonian Engine.

---

PAP

https://ift.tt/fbkFu4Z

Header Image Credit : Mbzt – CC BY 3.0

Adblock test (Why?)

YouTube adding AI dub technology to translate videos for creators - Dexerto - Translation

  ❘   Published: 2023-06-23T12:28:55

  ❘   Updated: 2023-06-23T12:29:05

A new feature for YouTube will allow creators to dub their videos into other languages with an upcoming AI integration.

At VidCon 2023, YouTube announced that it would begin letting creators dub their videos into other languages via the use of AI. The software, developed by Aloud and part of Google’s Area 120 incubator, will open up content to a much wider audience once implemented.

Speaking with The Verge, YouTube’s Amjad Hanif said that the tool is already being tested with hundreds of creators and that the AI program supports just a few languages right now.

Article continues after ad

Right now, you’ll find Spanish, Portuguese, and English available for translation purposes.

Currently, YouTube allows viewers to use automatic captions if the original uploader hasn’t provided subtitles. However, this can be very limited depending on the accents, and translating subtitles between languages is spotty or leads to worse accusations.

You can already try out the function yourself, as a science channel, Amoeba Sisters, seems to already have a video using it. To do so, just click the little gear, and audio tracks should be an option.

Article continues after ad

The results are very clearly AI-generated, with a robotic undertone as the Spanish voice tries to keep up with the English one.

Subscribe to our newsletter for the latest updates on Esports, Gaming and more.

YouTube introduces new AI dubbing tool for creators

Hanif mentions that the audio should improve in 2024, as Aloud continues to develop it. This will begin to mimic inflections, expressions, and lip sync.

While it’s unclear how YouTube would implement lip-synced video, we’ve already seen Nvidia alter the human body on the fly to keep eyes looking directly at the camera.

Article continues after ad

However, we suspect that it’ll function similarly to how YouTube is implementing the dubbing process. Aloud takes the audio, similar to the automatic captions. Once it has been processed and transcribed into text, you can edit and make adjustments. Finally, the AI voice is overlayed and integrated into the content.

Adblock test (Why?)

Inclusive Naming Initiative limps towards release of dangerous digital dictionary - The Register - Dictionary

The Inclusive Naming Initiative (INI) – an industry effort to promote and facilitate replacing harmful and exclusionary language in information technology – has developed its first recommendation lists, but has struggled to sign off on the announcement of the document.

Version 1.0 of the Inclusive Naming Word Lists offers three categories of lousy language worth losing:

• Tier 1 – Replace Immediately
• Tier 2 – Strongly Consider Replacing
• Tier 3 – Recommendations to Replace

The words and terms in Tier 1 are:

• Abort
• Blackhat/whitehat
• Cripple
• Master
• Master/slave
• Tribe
• Whitelist

As The Register has previously reported, several vendors have stopped using most of the terms in Tier 1 because of their racial connotations, or references to slavery.

We've not seen "tribe" on this sort of language list before. The INI wants it gone on grounds it "is a term that can be associated with colonialism or can be considered cultural appropriation" and "has a history of being used to describe division along ethnic or racial lines."

The INI suggests using other words to describe groups of people.

Only one term is listed in Tier 2: Sanity check. It earned its place on grounds it "might be derogatory to neurodiverse people." INI liked Twitter's preferred alternatives "confidence check" and "coherence check."

The INI's Tier 3 language list has just two entries, one of which is "Man-in-the-middle" because it is needlessly gendered and "implies that women do not have the skills to perpetrate this type of hacking."

The other Tier 3 term is "segregate." It's recommended for replacement on grounds that it "is almost exclusively associated with the segregation of Black people from White people, an extremely racist context."

INI has also defined some "Tier 0" terms that it discussed, but chose not to recommend for exclusion. Among the Tier 0 terms are "parent/child" which was felt not to describe an intentionally abusive relationship, "red team" because it's not a reference to indigenous people, and "fair hiring practice" which INI folk decided could not be construed as indicating a preference for light-skinned people.

"Blackbox" and "blackout" are also Tier 0 because they are "not based on a good/bad binary where white is represented as good or black is represented as bad."

While the v.1.0 list has been complete for some weeks, the INI's attempts to launch it formally appear to have stalled: the group's lightly trafficked forum contains a late May post asking for signoff of a draft announcement. At the time of writing no feedback had been added for almost four weeks, and one poster prodded the group to advance its release.

It's not the first attempt at cleaning the tech vocabulary of potential offence. Other initiatives have recommended discontinuing use of certain terms, with the likes of "hanging" and "killing" and "hit" stricken from some style guides. Others exclude "biting the bullet," "jumping the gun," and "taking a stab."

Cisco made the strangest recommendation we've found, suggesting you replace "kill two birds with one stone" with "feed two birds with one scone." Please note that it is inadvisable to feed most birds scones.

The INI's leaders include reps from IBM, Microsoft, Dell, Intel, Red Hat, Cisco, Extreme Networks, the Cloud Native Computing Foundation, and GitLab. ®

Adblock test (Why?)

Best summer books of 2023: Fiction in translation - Financial Times - Translation

What is included in my trial?

During your trial you will have complete digital access to FT.com with everything in both of our Standard Digital and Premium Digital packages.

Standard Digital includes access to a wealth of global news, analysis and expert opinion. Premium Digital includes access to our premier business column, Lex, as well as 15 curated newsletters covering key business themes with original, in-depth reporting. For a full comparison of Standard and Premium Digital, click here.

Change the plan you will roll onto at any time during your trial by visiting the “Settings & Account” section.

What happens at the end of my trial?

If you do nothing, you will be auto-enrolled in our premium digital monthly subscription plan and retain complete access for 65 € per month.

For cost savings, you can change your plan at any time online in the “Settings & Account” section. If you’d like to retain your premium access and save 20%, you can opt to pay annually at the end of the trial.

You may also opt to downgrade to Standard Digital, a robust journalistic offering that fulfils many user’s needs. Compare Standard and Premium Digital here.

Any changes made can be done at any time and will become effective at the end of the trial period, allowing you to retain full access for 4 weeks, even if you downgrade or cancel.

When can I cancel?

You may change or cancel your subscription or trial at any time online. Simply log into Settings & Account and select "Cancel" on the right-hand side.

You can still enjoy your subscription until the end of your current billing period.

What forms of payment can I use?

We support credit card, debit card and PayPal payments.

Adblock test (Why?)

Dhoomam movie review: This Fahadh Faasil starrer gets completely lost in translation - The Indian Express - Translation

Tobacco causes painful death! True. Equally true is the fact that watching a directionless film can also be a painful experience, although it may not result in someone’s death. However, what is even more disheartening is witnessing incredibly talented people waste their artistic abilities on a lacklustre endeavour.

Let’s consider the following scenario: a film directed by Pawan Kumar, known for his exceptional films such as Lucia (2013) and U Turn (2016), featuring a talented cast including Fahadh Faasil, Roshan Mathew, Aparna Balamurali, Vineeth, and Achyuth Kumar, and backed by the successful production house Hombale Films, known for their remarkable achievements with KGF and Kantara. Doesn’t this combination ignite excitement for the film’s release?

Sadly, their latest offering Dhoomam (Smoke) tragically falls short of expectations, despite its thought-provoking and relevant theme.

Set in Karnataka, Dhoomam tells the story of Avinash (Fahadh Faasil) and Diya (Aparna Balamurali), a couple plunged into deep trouble after Avinash decides to quit his lucrative position at a cigarette company.

Despite starting as the marketing head of the company, spearheaded by Sid (Roshan Mathew), and successfully leading numerous impactful campaigns which significantly boosted the firm’s sales, Avinash’s conscience begins to torment him when he realises that even children are falling prey to tobacco addiction, and the company willingly embraces these ‘new’ customers. This realisation compels him to leave his job.

Unfortunately, Avinash’s troubles escalate after submitting his resignation. Avinash and Diya are abducted by an unknown person who plants a time bomb inside one of them, forcing the couple to gather Rs 1 crore within a specific timeframe. The bomb’s timer can only be kept from ticking down by, ironically, smoking cigarettes. Failure to comply will result in an explosion. Seeking help from Sid and his uncle Praveen (Vineeth), Avinash finds himself ignored and neglected. However, the couple devises a plan to acquire the required fund through the company itself.

As their desperate journey unfolds, tragedy strikes when Praveen is shot dead and Sid is wounded by a bullet. Falsely accused of these crimes, Avinash and Diya become the prime targets of the police. The remainder of the film focuses on their relentless efforts to save themselves from this perilous predicament.

Advertisement

Dhoomam, as suggested by its title and trailer, revolves around the theme of cigarette smoking and its impact on countless lives. The film highlights how the heads of tobacco companies, fully aware of the health risks associated with their products, avoid consuming them. However, these very firms employ various tactics to promote cigarette use, leading to them becoming filthy rich while users gradually succumb to the clutches of death.

While a film that aims to deliver a significant message is commendable, considering the influential power of the medium, a movie solely created for this purpose can paradoxically yield counterproductive results. Dhoomam stands as a testament to this as it excessively prioritises the subject matter while neglecting the technical aspects.

In one scene, Avinash accurately summarises the film as a whole, stating, “You all must have seen the (anti-smoking) public service ads that are played in our movie theatres. What if we make such ads in a way that people will actually see it?” While Dhoomam can be seen as a 2-hour-24-minute-long public service advertisement, its ability to captivate audiences is questionable. The primary reason for this lies in the film’s weak script and poorly crafted dialogues, resulting in very little substance beyond its central theme.

Advertisement

Despite the sluggish pace of the initial act, one would assume it is building towards a significant impact. However, Dhoomam disappointingly offers very little in that regard, ultimately presenting a series of disconnected scenes that fail to make compelling points. Not only does the film lack emotionally resonant moments, but it also struggles to engage the audience throughout. Additionally, the frequent flashbacks, although intended to enhance the narrative, ironically hinder the viewing experience, as they focus only on how cigarette companies make strategies to boost sales. While these sequences had the potential to make a profound impact, they end up being mere recounts of Avinash’s past without offering substantial depth or substance to the overall story.

One of the major drawbacks of the film lies in its excessive reliance on dialogues. Dhoomam fails to let a single moment pass without characters resorting to explanatory dialogues, almost resembling a running commentary for the viewers. At the same time, the manner in which these dialogues are written indicates that much of their essence was lost in translation. The stark contrast between the Malayalam dialogues and the natural and flowing Kannada dialogues here, penned by writer-director Pawan Kumar himself, further emphasises this point. It demonstrates how the film could have fared much better if it had been made in Kannada and if the narrative style had been adjusted to align with contemporary trends, instead of the 2010s.

On the performance front, both Fahadh and Roshan showcase their talent impressively, both individually and in their scenes together. However, the contrived dialogues and drab script undermine their efforts, resulting in a somewhat childish portrayal overall. Despite her previous successful performances, Aparna Balamurali disappoints here in her depiction of a woman paying the consequences for her husband’s actions, as it comes across as unconvincing.

Although his role is minor and lacks a significant contribution to the film, Achyuth Kumar manages to leave a lasting impression. Joy Mathew and Nandu also deliver decent performances with the material they are given, while Vineeth’s acting fluctuates between impressive and unimpressive in alternating scenes.

In terms of technical aspects, Preetha Jayaraman’s cinematography and Anees Nadodi’s art direction shine in Dhoomam, capturing visually captivating scenes and creating a compelling atmosphere. However, Poornachandra Tejaswi’s music fails to make a lasting impact and leaves little impression on the viewers.

Advertisement

In short, Dhoomam serves as a reminder that merely having a significant message is insufficient to create a captivating film; it would, at best, help make a public service advertisement. To truly engage and captivate audiences, a film requires a multitude of other elements that should work harmoniously together.

Dhoomam movie cast: Fahadh Faasil, Aparna Balamurali, Roshan Mathew, Achyuth Kumar, Vineeth
Dhoomam movie director: Pawan Kumar
Dhoomam movie rating: 1.5 stars

Adblock test (Why?)