Call Transcription transforms the audio of your calls into speaker-organized scripts. This can be a great tool for reviewing calls quickly, spot-checking performance, and identifying trends. The cost of transcriptions varies from plan to plan and is billed at a per-minute rate.
Transcriptions are available for both inbound and outbound calls and are only compatible when call recording is turned on. To learn more about call recording, see the full article here.
With transcriptions enabled, you can use the following features:
- Speaker-aware text transcriptions will become available for transcribed calls in the call log, with text broken into sections based on who was speaking. Select sections of text in the transcription to jump to that part of the audio.
- Include transcriptions in notification emails to get a quick overview of a call or receive transcribed voicemails via email.
- Search the call log using words from transcriptions. If you remember a past conversation but can't remember the name of the person you spoke to, try searching in the call log for something you talked about.
- Keyword Spotting lets you identify words to look for and take action when those words are found in your transcribed calls.
Transcriptions are enabled in call settings. By default, each account has one call settings configuration ("Account Level") that all tracking numbers are assigned to. If you have only one configuration, these changes will apply to all tracking numbers on your account. You can create additional settings configurations to have different options enabled for different tracking numbers.
- Navigate to Numbers → Call Settings and click edit next to the call settings configuration where you'd like to enable transcriptions.
- Call recordings must be enabled for transcriptions to work. Click here to learn how to enable call recording in your account.
- Scroll down to the Transcription section and choose the country. Note: each country may have a different per-minute rate.
- Next, you have the option to choose Delay notifications until transcription is ready. Turning this on can add 1 to 2 minutes of post-call delay for transcription processing until email notification is delivered. By leaving this off, some emails will not include a transcription if processing has not finished when the email is sent.
- Live Transcription - Transcribes the audio in near real-time (Additional costs do apply, per minute at $0.02) Live transcription is not supported with call forwarding. You must be using a queue or dialing an agent. Note: redaction is not currently available with this feature currently.
- Maximum transcription length - (leave at 0 to transcribe the full recording) Billing will be rounded up to the nearest minute. Lengths below 10 seconds will be ignored. Using truncation with PCI, SSN, and PII detection is not possible.
- Natural Language Processing - Provides a summarized conversation and extracts entities e.g. classification of works such as Person, Quantity, Cardinal, Time, Etc.
- Enable Keyword Spotting to automatically classify and identify your calls. Enabling transcriptions also enables keyword spotting, which is included in the cost of transcriptions. Keyword spotting allows you to set up automatic actions that will take place when specific terms are found in the transcription of your call. For more information about keyword spotting, see the full article here.
- When enabling Redaction it will remove a portion of the call audio that includes sensitive PCI data and apply redacted and PCI tags automatically.
- Next select Save Changes
Transcription Security Features
Secure call transcriptions allow the system to detect when credit card information, social security information, or phone numbers are spoken during a call, tag the call appropriately, and redact that information from your call transcriptions and associated call recording.
For more information on secure transcriptions, see the full article here.
Transcriptions are available in the call log. When your transcription is processed, a page icon will appear in the Contact column, which you can click to bring up the completed transcription. You can also click edit on the left side of the call to open the detail panel, then click the Voice Analysis tab.
The transcriptions will be segmented by the speaker. You will see timestamps below each section of the speech, which will take you directly to that section of the call audio when clicked.