Speaking a foreign language often comes with fear of performance. What if you pronounce words wrongly? What if your mind gives up? Undoubtedly, it takes time to create confidence in speaking a language other than its origin, but the first step to reach there is really simple: practice.
In the world of language AppsBabbel is starting two speech-based tools to help its users using their oral skills, removing that fear and feeling comfortable while speaking.
The first new feature uses AI-operated speech recognition to help users during the vocabulary-based lessons and help them with pronunciation.
To evaluate the pronunciation of the speaker Aye The model was trained at millions of data points from the company’s audio bank, using both correct and wrong accents, different voice range and diverse pronunciation and dialects – for example, Spanish from Argentina, Colombia and Spain.

When activating the feature, users are asked to speak directly into their device when indicated. The AI ​​tool analyzes recording and compares with thousands of samples of phones (making separate sounds a Languga) to assess the accuracy of the accent. It then sewed users, provides visual response.
,[It] Works by comparing an effort with many possible positive results, with a forecast that can be fine according to the level of the learner (we can be strict with more advanced learners for example), “Product Marketing Manager Amin Guaraui in Babbel told TNW.


The second new speech-based tool of the app is everyday conversation, designed Help the learners practice dialogues with the goal of reducing the anxiety of speaking independently.
“Part of difficulty [in learning a language] While speaking comes down to confidence, “Molly Thompson Zugar, VP, VP, told TNW.
“The problem is that everyone wants to understand and it is understood to whom they are talking to, but practice to feel enough to do so. It is a catch 22 of a catch 22 and that is why we have launched everyday interaction facility.”
Every day the conversation simulates small dialogues between speakers in different real-life scenarios, such as “identity,” “ordering food,” or “talking to friends after work.”


Users first listen to conversations to get acquainted with it and then they can imitate it independently. According to the company, it creates a safe practice space, where the learners may feel comfortable to speak and enhance their oral skills.
Both features can be used individually or as additional support for learners who participate in live classes of Babbel, in line with the company’s language learning.
The AI ​​speech recognition is currently available on the web app in five languages: English, French, German, Italian and Spanish (Castilian and Latin American), planned to reach the widespread audiences in the coming months. It will soon be rolled out on iOS.
Meanwhile, everyday dialogue is available on Android and iOS for English and Spanish (Castilian and Latin American) learners. French will be added later in November – if you always dream of ordering a crocisain like a supporter.