The service provides conversion of recorded speech input to text output. It also provides identification of language in a recorded speech sample, and it can synchronize speech with corresponding text. It handles content in many European languages as well as Mandarin and Arabic. Geographic redundancy helps to provide effective service across countries and regions.
API methods support either real-time or batch processing of automated transcription jobs, from submission of audio sample through delivery of text. Methods also support basic detection of language spoken in audio samples and synchronization of audio with corresponding text transcript.