In recent times, numerous tools and services for transcription of voice-to-text conversations have been arriving, and although over time they have perfected precision, the same is not the case with voice recognition models.
This is where it now comes Bhasa, a new web-based tool, specialized in audio and video to text transcriptions. They understand that although automatic speech recognition technology has been around for a long time, as a general rule they still have problems understanding accents and are limited to a few languages.
His solution tries to fix the existing gap:
At Bhasa, we set out to fix this gap on both the technical and product fronts. Our speech recognition models are designed to work for the individual without excuses from accents and languages. Not forgetting, our biggest challenge and consequently the focus remains to create product experiences that deviate from technology and focus more on solving our customers’ problems.
Those interested can be part of the service, currently in beta, through its different pricing plans, with a first option for free. According to the capacities contracted through the different plans, users will be able to obtain the transcripts either by uploading a multimedia file, accepting mp3, mp4, wav files, or by providing a link to YouTube.
Regarding YouTube, they will be able to obtain the transcription so that they can also use it as subtitles for the videosAlthough users will be able to download the transcript in other file formats that allow it to be opened in a text editor for correction and improvement.
Users will be able to train Bhasa so that they can identify their voice, and they will even be able to train Bhasa to be able to identify multiple voices that participate in the same conversation.
According to Bhasa:
Bhasa’s audio-to-text and video-to-text conversion can be used in a number of ways. From dictations to conversations in professional and casual settings, here are scenarios where Bhasa can help you.