Transcription (ASR) — Overview
client.asr provides access to Decrackle's Automatic Speech Recognition capabilities.
Methods
| Method | Description |
|---|
client.asr.transcribe() | Transcribe an audio file (batch, synchronous) |
client.asr.transcribeStream() | Transcribe with real-time streaming output (Coming Soon) |
client.asr.jobs.list() | List past transcription jobs |
client.asr.jobs.get() | Get a single job by ID |
client.asr.jobs.delete() | Delete a job and its files |
client.asr.jobs.refreshUrls() | Refresh expired presigned download URLs |
Supported languages
| Code | Language |
|---|
en | English |
de | German |
fr | French |
es | Spanish |
it | Italian |
hi | Hindi |
ja | Japanese |
ar | Arabic |
Pass either the ISO code ("en") or the full name ("English"). Omit the language option entirely for automatic language detection.
MP3, WAV, FLAC, OGG, M4A, AAC
Plan requirements
| Feature | Free | Professional |
|---|
| Batch transcription | ✓ | ✓ |
| Quality scoring | — | ✓ |
| Streaming transcription (Coming Soon) | — | — |
Limits
| Limit | Free | Professional |
|---|
| Max file size | 10 MB | 100 MB |
| Max duration | 5 min | 120 min |
| Default timeout | 5 min | 5 min |