ReplicateSpeech to textAvailable now
Replicateturian/insanely-fast-whisper-with-video
turian/insanely-fast-whisper-with-video is a speech to text tool from Replicate that runs in the background and returns structured JSON. Pay as you go from $0.0012675 per second, with one API key.
Try itfrom $0.0012675 / second
Pricing
$0.0012675 per second of GPU time, measured on each run. The final charge comes from usage.
Limits
- Runs
- In the background
- Returns
- json
- Longest queue wait
- 3600s
- Run deadline
- 960s
- Largest inline input
- 1024 KB
- Largest input file
- 500 MB
- Largest output
- 4768 MB
- Outputs per run
- 16
Input
| Field | Kind | Default | Description |
|---|---|---|---|
url | text | — | Video URL for yt-dlp to download the audio from. Either this or audio must be provided. |
task | one of transcribe, translate | transcribe | Task to perform: transcribe or translate to another language. (default: transcribe). |
audio | URL or file | — | Audio file. Either this or url must be provided. |
language | text | — | Optional. Language spoken in the audio, specify None to perform language detection. |
timestamp | one of chunk, word | chunk | Whisper supports both chunked as well as word level timestamps. (default: chunk). |
batch_size | number (whole) | 64 | Number of parallel batches you want to compute. Reduce if you face OOMs. (default: 64). |
Same request in code
curl https://api.spiral.ai/v1/runs \
-H "Authorization: Bearer $SPIRAL_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
-d '{
"provider": "replicate",
"model": "turian/insanely-fast-whisper-with-video",
"input": {}
}'