Transcribe Speech to Text Node
Transcribe audio to text using Deepgram with speaker diarization, word-level timing, and detailed metadata.
Transcribe audio to text using Deepgram with speaker diarization, word-level timing, and detailed metadata.
Inputs
| Label | Property | Type | Description | Default Value | Is Required |
|---|---|---|---|---|---|
| Audio URL | url | botdojo/types/string | Publicly reachable URL to the audio file. | null |
Output
| Label | Property | Type | Description | Default Value |
|---|---|---|---|---|
| Transcript | transcript | botdojo/types/string | Plain text transcript of the audio | null |
| Duration (seconds) | durationSeconds | botdojo/types/number | Audio duration in seconds | null |
| Detected Language | detectedLanguage | botdojo/types/string | Detected language code (e.g., 'en') | null |
| Words | words | botdojo/types/data | Array of words with timing, confidence, and speaker info | null |
| Paragraphs | paragraphs | botdojo/types/data | Array of paragraphs with speaker diarization and sentences | null |
| Metadata | metadata | botdojo/types/data | Request metadata including request_id, duration, models used | null |
| Full Response | fullResponse | botdojo/types/data | Complete raw response from Deepgram API | null |