> For the complete documentation index, see [llms.txt](https://dataqueue.gitbook.io/voicehub-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://dataqueue.gitbook.io/voicehub-docs/agent-design-and-usage/analyze-past-calls-callstudio.md).

# Analyze Past Calls (CallStudio)

VoiceHub’s CallStudio allows you to upload and analyze past calls — even if they weren’t made inside VoiceHub — using the same AI engine that powers real-time call analysis.

### **Upload a Job**

1. Go to CallStudio from the sidebar
2. Click + Add Job
3. Give the job a title
4. Upload an audio file (MP3 or WAV, max size: 21MB)
5. Click Save

Once uploaded, VoiceHub will transcribe the file and perform AI analysis just like with a live call.

### **Review the Analysis**

<figure><picture><source srcset="/files/ZvcNENtDPQgFQtOB2bZJ" media="(prefers-color-scheme: dark)"><img src="/files/LbmBVP5BbUXoZD1s76Qe" alt=""></picture><figcaption></figcaption></figure>

After processing, each uploaded job generates detailed metrics:

* Transcript
* Total Call Duration
* Agent Duration vs. Client Duration
* Agent/Client Speaking Ratio
* Interruptions Count
* Record Summary (LLM-generated overview)
* Main Topics Timeline
* Detected Sentiment (Positive, Neutral, Negative)
* Resolution Status (Manually editable or AI-evaluated)
* Positive & Negative Keywords and Phrases
* Downloadable Audio (for listening to the call in full)

You can play back the call while reviewing its metrics — great for human QA, training, or validation.

Use CallStudio for:

* Post-call audits
* Manual call QA
* Human-in-the-loop scoring
* Uploading logs from external contact centers

This feature complements real-time Call Logs by providing deeper insights into legacy audio — especially useful during migration or model fine-tuning.