Uploading recordings
Already have a recording — a sales call, an interview, a support conversation? Upload it and Perceive8 produces the same transcript, scoring, and insights you get from a live meeting.
Upload a file
- Go to Audio → Upload.
- Drag one or more files onto the drop zone (or click to browse).
- Choose the language (or leave it on auto-detect).
- Pick a quality tier and any compliance options (see below).
- Click Upload.
You can safely navigate away — processing continues in the background. When it finishes, the result opens as an analysis and also appears under Audio → Analyses. Most recordings finish in a few minutes; longer files take proportionally longer.
Supported formats and limits
| Formats | WAV, MP3, M4A, OGG, FLAC, WEBM, MP4, MOV |
| Max file size | 500 MB per file |
| Duration | No explicit limit — very long recordings simply take longer to process |
Video files are fine: Perceive8 extracts and processes the audio track.
Languages
Auto-detect handles most cases well. You can also pin the language explicitly for better accuracy. Supported languages:
English, Spanish, French, German, Hebrew, Italian, Portuguese, Japanese, Chinese, Arabic, Russian, Korean, and Hindi.
Quality tiers
The tier controls how deep the analysis goes:
| Tier | What you get |
|---|---|
| Lite | Transcription and speaker labels |
| Pro | Everything in Lite, plus sentiment, entities, and topics |
| Max | Everything in Pro, plus automatic chapters and key phrases |
Higher tiers take a bit longer and consume more of your plan's usage. If you're unsure, start with Pro.
Compliance options
Two optional switches are available on the upload form:
- PII redaction — detects personal data (names, emails, phone numbers, IDs) in the transcript and masks it.
- Content moderation — flags toxic, abusive, or unsafe content in the conversation.
If your workspace handles regulated conversations (health, minors, finance), also review the consent model and Privacy & consent essentials.
Scoring the upload
If your workspace has an active playbook, it applies to uploads automatically — the analysis is scored against your rubric with no extra steps. The upload form shows which playbook is active, with a shortcut to Audio → Playbooks if none is set up yet. See Scoring with playbooks.
After the upload
While processing, the analysis shows a Processing status in Audio → Analyses (the list refreshes automatically). When it turns Ready, open it to explore the transcript, scorecard, topics, and reports — or ask Aria to summarize it for you. You'll also see a prompt to generate an Aria Report right after completion.
Tips for best results
- Use the cleanest audio you have. Speaker separation and accuracy track audio quality more than anything else.
- Pin the language for calls with heavy accents or code-switching.
- Name files meaningfully before upload — the file name becomes the analysis name (you can rename it later in the analyses list).