Audio Analysis

InsightsHub Audio Analysis is a tool designed to help research, customer success, sales, and call center teams manage and extract value from unstructured audio research assets, such as in-depth interviews, focus groups, customer calls, and meeting recordings. By ingesting audio files directly into an InsightsHub Repository, the system automatically transforms raw recordings into searchable, structured organizational knowledge through automated transcription, multi-language support, summaries, AI-generated insights attributed to specific speakers, theme-based tags, and QuestionPro AI Chat.

Watch this video to learn how to use Audio Analysis in InsightsHub:

You can ingest audio recordings into an InsightsHub Repository to begin the automated analysis process by following these steps:

• Log in to your QuestionPro account and navigate to InsightsHub.
• Select an existing repository or create a new one.
• Navigate to the folder where you want to store the audio files.
• Click the Import Files button to manually upload files from your device, or select a cloud connector to import directly from integrations such as Google Drive or SharePoint.
• Select your supported audio file (MP3, WAV, or M4A format). Ensure the file is at least 30 seconds long and does not exceed the 500 MB file size limit.

Once the upload is confirmed, InsightsHub automatically processes the audio content in the background. You will receive an in-app notification once transcription and AI analysis are complete.

Note: If you open the File Analysis view while processing is still underway, a progress bar will display status messages such as "Analyzing content" or "Drawing up conclusions." During this time, the Summary, Insights, and Tags tabs will remain locked until processing finishes.

The File Analysis view serves as your central workspace for reviewing and exploring audio research. It is structured into key functional areas:

• Audio Player: Located on the left side, enabling you to play, pause, and listen to the recording alongside your research data.
• Transcript Area: Displays the automatically transcribed text along with detected speaker labels and clickable timestamps. Clicking any timestamp jumps the audio playback directly to that exact moment in the recording.
• Tabbed Analysis: Located on the right panel, allowing you to toggle between the Summary, Insights, and Tags tabs to access AI-generated synthesis and organization tools.

The Summary tab provides a quick, structured brief of long audio recordings without requiring you to listen to the full file.

As soon as audio processing and transcription are finished, an AI summary is generated automatically.

• Recap: Read a concise natural language overview highlighting the primary discussion topics and findings from the recording.
• Actions: Review key takeaways, next steps, or strategic recommendations extracted by AI from participant dialogue.

You can copy the summary text to share directly with stakeholders or provide feedback on summary accuracy to help train and refine future AI performance.

The Insights tab helps you surface critical evidence and key quotes from your qualitative audio files without manual searching.

• Navigate to the Insights tab within the File Analysis view.
• Select Suggest Insights to prompt the AI to identify significant moments and quotes across the transcript.
• Review each suggested insight, which includes an accompanying timestamp, quote segment, and speaker attribution.
• Accept or Reject: Click to approve individual suggestions or bulk select insights to commit them to your repository project.

Accepted insights display metadata indicating the creation source (AI or User), date committed, and the identified speaker. You can also use QuestionPro AI Chat at the repository level to ask natural language questions about your uploaded audio data and receive source-backed responses.

The Tags tab offers a centralized view of key themes and sentiment detected across your audio assets.

• Theme and Sentiment Tags: View automated tags generated by AI alongside manual tags applied from the transcript menu.
• Tag Counts: Review how frequently specific tags occur across your repository insights.
• Centralized Exploration: Click any tag to display all related insights, verbatim audio quotes, original timestamps, speaker metadata, and associated details.

InsightsHub supports MP3, WAV, and M4A audio file formats. Audio recordings must be at least 30 seconds long. There is currently no maximum duration limit, subject to the 500 MB file size limit.

Yes. InsightsHub automatically detects the language of the audio during transcription. Audio processing is not restricted to English.

InsightsHub automatically detects and distinguishes speakers in an audio recording. The transcript is organized by speaker, with timestamps that allow you to jump directly to the corresponding point in the recording. Distinct speakers detected in the recording are captured as part of the analysis, and AI-generated insights are attributed to the speaker associated with the relevant statement.

You can upload up to 50 audio files in a single manual batch upload.