Skip to content
AskVault Docs
Try Free →

Voice transcript routing

Written by , Founder, AskVault · Reviewed by Aashiq
Last updated: · 4 min read

How transcription works

During and after each call:

  1. Twilio streams audio to AskVault.
  2. ASR (automatic speech recognition) transcribes in real time.
  3. Bot analyzes for intent, sentiment, named entities.
  4. Routing rules apply post-call: assign team, tag, escalate.
  5. Transcript available in the conversation view within 60 seconds.

Word-level accuracy: about 85 to 95% on clear English. Lower for noisy audio or strong accents.

Routing patterns

Three common patterns:

By topic. Transcript mentions "refund" → route to billing team. Mentions "down" or "outage" → route to engineering.

By sentiment. Negative sentiment detected → assign senior agent immediately. Positive sentiment + question resolved → auto-close.

By caller identity. Verified Enterprise customer → route to dedicated AE. Anonymous or free-tier → route to general support.

Configure under automation rules using transcript content as the condition.

Sample routing rule

For an SaaS:

  • Trigger: voice call ended.
  • Conditions: transcript contains "cancel" OR "refund".
  • Actions: tag billing, assign to billing-pool, notify #billing-alerts in Slack.

Apply rules per call. Routing typically completes within 5 minutes of call end.

Transcript structure

Per call:

  • Full transcript with speaker labels (Caller / Bot / Agent).
  • Timestamps per utterance.
  • Confidence scores per word.
  • Detected entities (names, dates, order numbers).
  • Sentiment trajectory across the call.

Visible under Conversations > [call] > Transcript.

Searching across transcripts

Like text conversations, voice transcripts are full-text searchable:

  1. Conversations > Search.
  2. Filter channel = voice.
  3. Type the search query.
  4. Results show matching call segments with timestamp.

Useful for QA, training, and pattern analysis.

Multi-language transcription

Auto-detection plus explicit selection:

  • Auto-detect from first utterance. Switches mid-call if caller switches.
  • Workspace default language if auto-detect fails.
  • 35 languages supported.

Accuracy varies:

  • English. 90 to 95%.
  • Spanish, French, German. 85 to 90%.
  • Hindi, Mandarin, Japanese. 80 to 88%.
  • Less-resourced languages. 70 to 82%.

Speaker diarization

For multi-party calls (caller + bot + transferred-agent):

  • Each speaker labeled distinctly.
  • Transcript shows who said what with timestamps.
  • Audit-ready for compliance.

Accuracy on 2-party calls: about 95%. Drops on 3+ parties.

Privacy and PII redaction

For sensitive content:

  • Optional automated redaction of credit cards, SSNs, account numbers from transcripts.
  • Compliance with PCI scope when redaction is enabled.
  • Audit log of redactions performed.

Enable under Voice > Transcription > PII Redaction. Recommended for regulated industries.

Call recording retention

  • Audio recordings retained 90 days standard, 1 year Enterprise.
  • Transcripts retained for conversation's full retention (365 days standard).
  • Configurable per workspace.

Useful for legal hold or extended training.

Integration with downstream tools

Send transcripts to other systems:

  • Webhook on call end. Push full transcript to your CRM.
  • Zendesk ticket creation. Bot creates ticket with transcript as body.
  • HubSpot activity log. Transcript attached to contact record.

Configure under automation rules with webhook actions.

Analytics

Per workspace:

  • Calls per day.
  • Average call duration.
  • Bot-resolved vs human-resolved split.
  • Most common topics (from transcript clustering).
  • CSAT if surveys enabled.

Visible under Analytics > Channel: Voice.

Limits

  • Transcription latency. Real-time during call; final transcript within 60 seconds post-call.
  • Max call duration. 2 hours; longer calls split.
  • Recording size. Variable; about 1 MB per minute of audio.
  • Routing rules per workspace. 25 (Growth), 100 (Business).

Common pitfalls

Transcription quality low. Background noise or poor audio. Improve caller audio (request hands-free or quiet environment).

Wrong routing. Conditions too broad. Tighten with multi-word phrases.

PII not redacted. Redaction off. Enable under Voice settings.

Transcript missing words. ASR limit. Cross-check with audio recording.

FAQ

Can I export transcripts as CSV?

Yes via API or Conversation Export. Includes timestamps and confidence scores.

Does transcription work for voicemail?

Yes. Voicemail audio transcribes within 60 seconds.

Is the transcript searchable in the customer's conversation history?

Yes. Cross-channel search treats voice transcripts like any other message.

Was this page helpful?