Audio Reference
Narration settings, generation limits, dictionary limits, and delivery behavior.
Narration settings
| Setting | Values | Default |
|---|---|---|
| Voice | Available English or Spanish voice | Professional narrator |
| Generation speed | Slow: 0.85x, Normal: 1x, Fast: 1.15x | Normal |
| Pronunciation dictionary | A synced dictionary in the active scope | None |
| Playback speed | 0.75x, 1x, 1.25x, 1.5x, 2x | 1x |
Voice availability can change. The library groups voices by presentation style and language. Voice labels can be masculine, feminine, or neutral.
Generation
Y2 removes citation markers, the trailing references section, HTML, and Markdown before narration. It converts numbers, abbreviations, and punctuation for speech. Assigned branding supplies the intro, outro, and optional before-report and after-report content.
| Property | Value |
|---|---|
| Provider model | Cartesia Sonic-3 |
| Format | MP3 |
| Sample rate | 44.1 kHz |
| Bit rate | 128 kbps |
| Maximum file size | 20 MB |
| Duration before playback | Estimate from the prepared text |
Provider configuration, an incomplete report, file size, or report budget can prevent generation. Audio is optional. Delivery can continue without it.
Dictionary limits
| Property | Limit |
|---|---|
| Dictionaries in Pro | 10 |
| Dictionaries in Elite | 100 |
| Entries per dictionary | 200 |
| Dictionary name | 50 characters |
| Description | 200 characters |
| Source word | 100 characters |
| Replacement | 300 characters |
Spoken branding
| Field | Limit | Position |
|---|---|---|
| Audio intro | 100 characters | Before the report |
| Before-report audio | 200 characters | After the intro and optional label |
| After-report audio | 200 characters | Before the outro and optional label |
| Audio outro | 200 characters | At the end |
Use {name} for the profile name. A branding template must be assigned to the profile.
Delivery
| Surface | Behavior |
|---|---|
| Report page | Shows Listen when an audio URL exists |
| Links to the report page without attaching the MP3 | |
| Webhook | Supplies availability, estimated duration, and an API audio link |
| Reports API | Returns audio metadata or redirects to the MP3 |
Deleting the owning report, profile, workspace, or account removes its audio. Storage cleanup can also remove files. Save a copy if you need an archive.
API fields
Use durationSeconds. The older duration field is deprecated.
Request ?redirect=true or an audio media type for a 302 redirect to the stored file.
The audio-text endpoint requires reports:read.
It returns prepared speech text and estimated duration even when no MP3 exists.
See the audio endpoint reference for the complete schema.