Sherlock Calls
for ElevenLabs + VAPI
VAPI runs your AI agents. ElevenLabs provides their voice. Sherlock Calls connects both — so you can investigate which voices make agents more effective, track synthesis costs per conversation, and diagnose audio issues that affect outcomes.
TL;DR — What beta users get access to
- 1
Sherlock Calls connects to both ElevenLabs and VAPI with read-only API access, letting you investigate the relationship between voice synthesis quality and agent conversation performance from a single Slack question.
- 2
Ask questions neither platform can answer alone: which ElevenLabs voices produce the highest VAPI conversation completion rates, what the combined TTS + agent cost per successful call is, and which voice configurations minimize early hang-ups.
- 3
Sherlock correlates ElevenLabs usage data with VAPI conversation outcomes and returns a sourced answer in under 5 seconds.
<5s
Voice + agent answer from Slack
2
Voice AI platforms, 1 question
0
Code or dashboards required
The Investigation Gap
What's invisible when you use ElevenLabs + VAPI without Sherlock
Each platform shows you its own data. But the questions that matter most live in the gaps between them.
Voice quality impact on conversation success is unmeasured
VAPI tracks conversation outcomes. ElevenLabs tracks synthesis parameters. But whether one voice model produces better engagement than another — measured by real conversation completion rates in VAPI — is data that requires correlating both platforms.
Combined TTS + agent cost per conversation is invisible
ElevenLabs bills per character synthesized. VAPI bills per minute of agent time. Your actual cost per completed conversation spans both — but you're only ever seeing half the cost picture in each platform's dashboard.
Audio-related early hang-ups aren't attributed to synthesis
When VAPI logs a conversation that ended in the first 5 seconds, it could be a wrong number — or it could be a voice quality issue. Without correlating ElevenLabs synthesis parameters for that specific call, you can't tell the difference.
Cross-Provider Questions
What teams ask Sherlock about ElevenLabs + VAPI
Questions that would take hours to answer manually — answered in under 5 seconds from Slack.
- SC“Which ElevenLabs voice IDs have the highest conversation completion rate in VAPI?”
- SC“What's our blended cost per completed conversation across ElevenLabs and VAPI this month?”
- SC“Show me all VAPI conversations that ended in under 10 seconds — what ElevenLabs voices were they using?”
- SC“Which ElevenLabs voices correlate with the longest VAPI conversation durations?”
- SC“Compare conversation sentiment scores in VAPI across different ElevenLabs voice configurations”
Beta Setup
Connect ElevenLabs + VAPI to Sherlock in 2 minutes
No code, no webhooks, no new dashboards. Beta users get direct onboarding support.
- 1
Connect your ElevenLabs account
Add your ElevenLabs API key to Sherlock Calls. Sherlock will index your voices, usage history, and synthesis logs.
- 2
Connect your VAPI workspace
Add your VAPI API key to Sherlock Calls. Sherlock will index your agents, call logs, and conversation outcomes.
- 3
Ask your first cross-provider question
Type 'Which ElevenLabs voices produce the best VAPI completion rates?' in Slack. Sherlock correlates both and returns a sourced answer.
FAQ
Common questions about Sherlock + ElevenLabs + VAPI
How does Sherlock correlate ElevenLabs and VAPI data?
- Sherlock matches ElevenLabs synthesis sessions to VAPI conversation records by call ID, timestamp, and voice configuration identifiers. It queries both APIs simultaneously and joins the results to produce unified answers.
Can Sherlock tell me which ElevenLabs voice converts best in my VAPI agent?
- Yes. Sherlock can rank your ElevenLabs voices by VAPI conversation completion rate, average duration, or any other conversation outcome metric — giving you data-driven guidance on voice selection.
Does Sherlock track combined ElevenLabs + VAPI costs?
- Yes. Sherlock pulls cost and usage data from both platforms and can compute blended cost per conversation, per voice, or per agent configuration.
What if I use multiple ElevenLabs voices across different VAPI agents?
- Sherlock handles multi-voice, multi-agent configurations. You can ask questions scoped to specific voice IDs, agent IDs, or any combination, and Sherlock will filter both datasets accordingly.
Is the ElevenLabs + VAPI combination available now?
- This combination is currently in invite-only beta. Apply for early access and we'll reach out if your stack is a good fit.
Apply for early access to Sherlock + ElevenLabs + VAPI
We're accepting a select group of beta users to shape the ElevenLabs + VAPI combination. Tell us about your stack and we'll reach out personally if you're a fit.
Explore individual integrations