Our client needed a trustworthy partner to transcribe and categorize a large quantity of high-impact audio data for voice recognition systems across ten different language markets. To accomplish this, we

  • Recruited and trained a team to complete the project in ten markets, each consisting of millions of audio files
  • Set up three highly secured offsite delivery centers in China, the U.S. and the U.K. within three weeks


We were able to securely ramp to full production in less than three weeks. In all ten markets, the data we gathered achieved 98% accuracy for grammar, syntax and para-linguistic tag use, as well as 98% accuracy for annotation.