Components

Description

File Directory

• Ensure a single, consistent directory where Sprinklr configuration searches for files.

• Ensure a one-to-one mapping between a specific importer configuration and a directory. No two importer configurations can point to the same directory.

File Extension and Content

​

• Sprinklr currently supports only zip or tar.gz files for voice data.

• Each zip file should contain one metadata file and multiple audio call recordings.

• The metatdata CSV or Excel in the zip file can have multiple rows.

File Prefix

​

• Ensure a consistent prefix for zip files. Examples of configuration searches for zip files with this consistent prefix are Partner_Call_1.zip, and Partner_Call_2.zip.

  ​Note: Inside zip or tar file, if Sprinklr does not find any metadata or recording
  file with expected extensions, the entire zip/tar file gets ignored.

  ​

Metadata File Extension

​

• Sprinklr currently supports three types of metadata file extensions: XLSX, XLS, and CSV. You must adhere to these extensions for all selected files.

Metadata Contents

Metadata file contains all case level details. Here are the fields present inside the file:

• unique_call_id: Refers to the ID of the particular Case. (Mandatory Field)

• call_start_time: Refers to the time when the Case started. If the timestamp of voice data is not present in the metadata, the current time of upload is considered in the case creation time.

• customer_id: Identifies the customer in a Case.

  Note: The system does not currently support the import of customer profiles. If the customer ID is present, a new profile is created; otherwise, the case is associated with a backend-created profile consistent across customers.

• call_language: Language in which the recording is provided.

• Agent_Id: Refers to the Agent handling the Case.

• call_direction: Specifies an outbound or an inbound call.

• Custom fields: The user can map custom field from the source system to Sprinklr custom fields.

• Recording_filename: Refers to unique name required to idenify a recording in the zip file. If the audio recording is traced inside the zip file, it gets attached to the unique_call_id and a case is created, else a case is created without the audio recording.

  Note: A Case is created irresepective of whether the audio file is present or not. In case the audio recording is not present, a case is created using only the metadata.

​​

Audio Files

1. Here are the different supported Audio formats: Stereo, WAV, MP3, OPUS(with a frequency of 8000 Hz or more), aac, wma, ogg, m4a, and mp4.

2. You can choose any one of the Pulse Code Modulation(PCM)) Formats:

   • Pcm_s16le

   • PCM A-law

   • PCM mu-law

   • G.729a

3. Optimizing for ASR Models:

   • Ensure audio files are representative and mimic actual production data and use cases for optimal Automatic Speech Recognition(ASR) model performance.

   • Include recordings with different speakers, providing a good distribution for both male and female genders.

   • Include sample calls/audios for each intent/use case.

   • Ensure availability of different speakers from various locations or with different accents.

Processing Time

• On an average, it takes around five seconds to create a case when there is a five minute conversation between a customer and an agent via SFTP.