Transcriptions

Post-Call or Standalone

Choose between post-call transcriptions or standalone transcriptions. Post-call transcriptions can be generated automatically after each call. Standalone transcriptions are generated on-demand for uploaded audio files.

Keyword Spotting:

Enable automatic classification of calls based on specific keywords detected in the transcription. Scan transcriptions for predetermined terms and trigger actions when these terms are identified.

PII (Personally Identifiable Information) Redaction:

Protect sensitive information by automatically redacting PII from transcriptions. Redaction types include: Personal Details, Government Issued IDs, Contact, Financial, Medical, Online Identifiers, Occupation & Professional, and Events & Dates.

Actionable Insight

Select a set of actions and webhooks that will be fired when keywords are spotted. This works for both post-call and standalone transcriptions.

Example Transcription

Transcriptions are displayed as a timestamped, speaker-labeled conversation log. Each entry shows the speaker (e.g., "Agent" or "Caller"), the timestamp within the recording, and the transcribed text. When PII redaction is enabled, sensitive data is replaced with category labels such as [PHONE_NUMBER] or [SSN].

Example API Request

# Submit standalone transcriptions as JSON:
curl -H "Authorization: Basic BASE64_ENCODED_PUBLIC_KEY_AND_PRIVATE_KEY" \
     -H "Content-Type: application/json" \
     -X POST \
     -d '{"caller_id":"+18004506787","file_url":"http://example.com/publicly_accessible_file.mp3","postback_url":"http://example.com/postback_url_without_authentication","data":{"loan_amount":"501","custom_token":"the_value"}}' \
     "http://[your-subdomain].trackdrive.com/api/v1/standalone_transcriptions"