OpenAI Whisper large-v3-turbo hosted by Cloudflare Workers AI, with word-level timestamps and automatic language detection.
12 languages · benchmarked
Whisper Large V3 Turbo is a speech-to-text model from Cloudflare. In our standardized benchmarks it reaches 12.90% word error rate, ranking #10 of 37 models tested. It supports 12 languages and runs at $0.001/min — the cheapest in the catalog.
Its nearest benchmarked alternative is MAI-Transcribe-2 (preview). Best suited for accuracy-critical work and high-volume, cost-sensitive pipelines.
Overall Score
76.7
#10 of 37
Word Error Rate
12.90%
Character Error Rate
9.21%
Match Error Rate
11.87%
Word Info Lost
16.31%
Avg Latency
5.3s
Benchmarks Run
33
overall WER · 37 models
Whisper Large V3 Turbo
field · 36 models
lower-left is better
8 categories
WER
lower is better
Medical
2.1%
Legal
2.8%
Technical
5.2%
Finance
6.7%
Conversational
10.0%
General
10.7%
Code-Switching
23.2%
Noisy Environment
36.3%
| Category | WER | CER | MER | WIL | Latency | Benchmarks |
|---|---|---|---|---|---|---|
Code-Switching | 23.15% | 14.58% | 22.50% | 30.97% | 8.1s | 2 |
Conversational | 9.99% | 6.79% | 9.97% | 11.77% | 3.0s | 2 |
Finance | 6.67% | 3.09% | 6.55% | 10.41% | 11.1s | 2 |
General | 10.65% | 8.56% | 9.76% | 14.50% | 5.1s | 17 |
Legal | 2.75% | 1.95% | 2.74% | 3.42% | 4.7s | 2 |
Medical | 2.10% | 0.99% | 2.10% | 3.22% | 7.2s | 2 |
Noisy Environment | 36.25% | 24.38% | 31.97% | 39.63% | 2.9s | 4 |
Technical | 5.17% | 3.03% | 5.17% | 6.76% | 3.8s | 2 |
5 accents
| Accent | WER | Benchmarks |
|---|---|---|
African | 24.62% | 1 |
Australian | 36.92% | 1 |
Indian | 23.81% | 1 |
British | 5.97% | 1 |
American | 9.43% | 1 |
Model ID
cloudflare/whisper-large-v3-turbo
Authenticate every request with your secret API key as a Bearer token. Issue a key from your dashboard.
Use it with your agent
Paste this into Claude Code, Cursor, Codex or any agent with a shell. It installs the CLI and the OpenTranscription skill, then transcribes with this model.
Quickstart
Upload your audio, create a job with this model, then poll the job or set a webhook_url to be notified when it completes.
Key parameters
file_path
string
required
Storage path returned by the upload step.
model
string
required
The model to run this job on.
models
string[]
Ordered fallback chain: primary first, then backups tried in order if a provider fails.
language
string
ISO 639-1 language code. Omit to auto-detect.
word_timestamps
boolean
Per-word start, end and confidence. Defaults to true; send false to leave them out.
webhook_url
string
HTTPS URL notified with the result when the job completes. Delivered events are signed (X-OT-Signature).
title
string
Display name for this job in the dashboard. Falls back to the file name.
Response
A completed job returns the transcript in the OpenTranscription Unified Schema (OTUS).
Webhooks — skip polling
We POST a signed event to your webhook_url when a job completes or fails; verify the X-OT-Signature (HMAC-SHA256, reject if older than 300 s) and dedupe on the event id, then fetch the full transcript via GET /api/v1/transcriptions/{id}.
Rate-limited per tier — see the X-RateLimit-* response headers.
Full API referenceUptime · 30d
100.00%
Error rate
0.00%
Avg latency
78.3s
Last incident
None in 30d
from the same field
same benchmark, every match
Supported Languages
Supported Formats
Features
Limits
Max file size
25 MB
Max duration
No documented limit