> ## Documentation Index
> Fetch the complete documentation index at: https://docs.smartinterview.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Qualitative Interview Analysis

> Upload audio or video recordings, get speaker-labeled transcripts with timestamps, and query your interview data using the built-in AI workspace chat.

Pulse Qualitative turns your interview recordings into searchable, structured data. Upload an audio or video file, choose your transcription mode, and Smartinterview returns a timestamped transcript with each speaker automatically identified. Once transcribed, you can chat with your recordings using plain language, no manual tagging or export required.

Use **Pulse Qualitative** for in-depth interviews, focus group recordings, user research sessions, or any qualitative data where understanding who said what matters.

## Key capabilities

* **Speaker diarization**: the advanced transcription mode automatically identifies and labels each participant (Interviewer, Respondent 1, etc.) so you can read or filter by speaker without manual editing.
* **AI workspace chat**: ask questions about your transcriptions in natural language and get answers grounded in the actual recordings.
* **Transcript export**: download the transcript as plain text or a Word document for use in reports or external analysis tools.
* **Multi-file workspaces**: group recordings from the same study into a workspace and chat across all of them at once.

## How it works

<Steps>
  <Step title="Upload your recording">
    Drag and drop one or more audio or video files onto the **Pulse Qualitative** upload area. Smartinterview accepts MP4, MOV, WebM, MP3, WAV, M4A, OGG, and other common formats. Files are compressed automatically before processing.

    [Learn about uploading →](/pulse-qualitative/upload-interviews)
  </Step>

  <Step title="Transcribe with speaker labels">
    Select the recording language and choose between **Standard** (faster) or **Advanced** (speaker diarization) transcription. Click **Transcribe** and monitor progress in real time.

    [Learn about transcription →](/pulse-qualitative/transcription)
  </Step>

  <Step title="Review the transcript">
    Open the completed transcript to read timestamped speaker turns, browse Q\&A pairs extracted from the conversation, and download the result in your preferred format.
  </Step>

  <Step title="Chat with your data">
    Open the **Workspace Chat** panel and ask questions about one or more transcriptions. Use natural language — for example, "What were the main concerns about pricing?" or "Summarize what Respondent 2 said about onboarding."

    [Learn about workspace chat →](/pulse-qualitative/workspace-chat)
  </Step>
</Steps>

## Token usage

Transcription costs **8 tokens per minute** of audio. To estimate the cost for a recording, divide the duration in minutes by one and multiply by 8. A 45-minute interview, for example, uses 360 tokens. You can top up your token balance at any time from your account settings.

<Note>
  Token charges apply to transcription only. Reviewing transcripts, browsing history, and using workspace chat do not consume tokens.
</Note>