Audio data and accuracy

To get the most from Voci's Automatic Speech Recognition (ASR) solutions, it helps to understand the variables that affect transcription accuracy. Audio is the primary input to an ASR system; therefore, the quality of the audio files have a significant impact on transcription accuracy.

In general, the best audio recording practices are to:

  • Use dual-channel (stereo) audio.

  • Use a sampling rate of 8kHz.

  • Use high quality telephony and recording equipment.

  • Minimize surrounding noise in environments where audio is recorded.

  • Record high quality audio with codecs that are optimized for speech.

  • Avoid lossy transcoding and compression, as with audio in MP3 format

The pages in this section elaborate on these audio recording practices.