About the Dataset

2006 hours

There are 1800 hours of Dutch (Netherlands) human-to-human audio with the following domain distribution per dataset:

• 482.17 hours of Banking
• 420.43 hours of Insurance
• 415.4 hours of Retail
• 475.53 hours of Telecommunication
• 7 hours of Generic data without specific domain

There are 206 hours of Dutch (Belgium) human-to-human audio with the following domain distribution per dataset:

• 58.6 hours of Banking
• 53.65 hours of Insurance
• 48.12 hours of Retail
• 46.37 hours of Telecommunication

Defined.ai creates scenarios for our crowd members to follow, which they study beforehand. They then record a conversation, one speaker playing the agent, the other speaker “playing out” the scenario with spontaneous content. The recording is done via telephony and is saved in 8khz 16 bit per channel. That content is then transcribed.

The dataset is covered by Defined.ai's standard license agreement. The license agreement is perpetual and allows for the commercialization of all models built on the data.

Other characteristics:

• Audio format: WAV
• Recording environment: noisy, silent
• Bits per sample: 16
• Communication band: broadband
• Sample rate: 8Hz

Metadata Distribution

Dutch Netherlands

Dutch Belgium

Samples

• 5-minute audio sample of Dutch (Netherlands). Transcription for the sample is also available.
• 5-minute audio sample of Dutch (Belgium). Transcription for the sample is also available.