Dosmono dictation takes advantage of the powerful AI technology of Dosmono, and relies on its core speech recognition technology to provide users with the recognition ability of Chinese, English, Japanese, Korean and many other languages to convert continuous audio into text as an hour audio can be transliterated within 5 minutes. The audio with standard pronunciation and clear voice can be transliterated into words as above 95% of accuracy rate and have an excellent experience for voice transliterate to word. Dosmono recording assistant terminal products (APP side, WEB website) provide users with product experience of one-stop transcription services such as voice transcriptions, audio transcriptions, real-time recording transcriptions, external audio import transcriptions, opening up Dosmono input method, associating users’ dictionary base, personalized transcription, dialogue translation, simultaneous interpretation, text intelligent editing, multiterminal synchronous editing, and voice cloud storage service.
1. Please provide audio files with standard pronunciation and no background noise as far as possible because the factors like pronunciation or background noise will affect accuracy rate of transliteration
2. Single audio size is no more than 2G, duration is no more than 5 hours.
3. Audio transliteration will compress the original audio in a unified format, please keep the original audio files if necessary.
Support wav、flac、ogg、amr audio format upload only support avc and avo file to upload which are generated by recording pen C1 and C1Pro of Dosmono
There are several situations in which a successful uploaded audio may not be found
1. After the success of audio uploads, the platform save them temporary, it will lose efficacy after a certain limited time. Please submit and pay the transliteration order after a successful audio upload
2. Audio will be deleted synchronous on the other side and cause the audio cannot be found on the website.
During the transliteration process, there is a very small probability that audio files will fail to be transcribed due to damage during uploading, decoding error, network and other reasons. Platform will give the corresponding audio payment duration resources back to the account after failed transliteration. You can check residual available duration resources. For failed audio transliteration, users can re-upload the audio and try to transliterate again.