We’re Israel & Mario, cofounders of Neuraan (https://neuraan.com). It is difficult for natural language processing developers in LatAm to get their applications to have the same accuracy in Spanish as in English. We create an API that helps them correct those inaccuracies and increase the adoption of their solution.
We have +6 years of experience developing chatbots and voice assistants for large enterprises across LatAm and was very common to end up doing some research and developing algorithms to enhance the accuracy that the commercial solutions like Dialogflow, Watson, Amazon Lex, and Azure Cognitive deliver because they provide a general solution for non-English languages.
To provide an extra layer to improve the performance of conversational applications in Spanish we developed a RESTful API that allows developers and enterprises to provide a few examples of the intents and entities used by their applications to enhance the recognition. We are using a few shot learning strategies developed using various Machine Learning strategies including deep neural networks along with self-developed algorithms.
This solution permits the enhancement of the recognition of goal-oriented applications (create a support ticket, extract data from a user conversation, make a telephonic sale, etc.) instead of standard NLP engines' general Spanish.
Our pricing is based on API requests.
We look forward to feedback from the HN community!
We’re Israel & Mario, cofounders of Neuraan (https://neuraan.com). It is difficult for natural language processing developers in LatAm to get their applications to have the same accuracy in Spanish as in English. We create an API that helps them correct those inaccuracies and increase the adoption of their solution.
We have +6 years of experience developing chatbots and voice assistants for large enterprises across LatAm and was very common to end up doing some research and developing algorithms to enhance the accuracy that the commercial solutions like Dialogflow, Watson, Amazon Lex, and Azure Cognitive deliver because they provide a general solution for non-English languages.
To provide an extra layer to improve the performance of conversational applications in Spanish we developed a RESTful API that allows developers and enterprises to provide a few examples of the intents and entities used by their applications to enhance the recognition. We are using a few shot learning strategies developed using various Machine Learning strategies including deep neural networks along with self-developed algorithms.
This solution permits the enhancement of the recognition of goal-oriented applications (create a support ticket, extract data from a user conversation, make a telephonic sale, etc.) instead of standard NLP engines' general Spanish.
Our pricing is based on API requests.
We look forward to feedback from the HN community!