Google Latest (Short) is a speech-to-text model from Google Cloud. In our standardized benchmarks it reaches 57.82% word error rate, ranking #35 of 35 models tested. It supports 40 languages and runs at $0.016/min — a premium option.
Its nearest benchmarked alternative is Soniox STT (Async). Best suited for multilingual workloads.
Overall Score
21.1
#35 of 35
Word Error Rate
57.82%
Character Error Rate
52.44%
Match Error Rate
57.66%
Word Info Lost
61.57%
Avg Latency
11.3s
Benchmarks Run
34
last 30 runs
Word error rate
57.82%
↑ 0.5% · 30d
3 runs ago
latest
Avg latency
11.3s
↓ 6.5% · 30d
3 runs ago
latest
overall WER · 35 models
Google Latest (Short)
field · 34 models
lower-left is better
8 categories
WER
lower is better
Medical
42.2%
Technical
44.7%
Legal
47.5%
Finance
47.8%
General
52.0%
Conversational
56.6%
Noisy Environment
89.5%
Code-Switching
97.3%
| Category | WER | CER | MER | WIL | Latency | Benchmarks |
|---|---|---|---|---|---|---|
Code-Switching | 97.26% | 91.98% | 96.34% | 98.43% | 16.7s | 2 |
Conversational | 56.58% | 50.98% | 56.14% | 61.35% | 11.7s | 2 |
Finance | 47.75% | 42.82% | 47.40% | 51.12% | 16.9s | 2 |
General | 52.01% | 46.38% | 51.95% | 56.72% | 10.1s | 18 |
Legal | 47.48% | 45.15% | 47.48% | 49.39% | 14.1s | 2 |
Medical | 42.18% | 39.47% | 41.69% | 45.45% | 12.0s | 2 |
Noisy Environment | 89.49% | 80.99% | 89.49% | 90.82% | 6.4s | 4 |
Technical | 44.66% | 41.68% | 44.56% | 48.89% | 16.7s | 2 |
5 accents
| Accent | WER | Benchmarks |
|---|---|---|
African | 84.62% | 1 |
Australian | 80.00% | 1 |
Indian | 90.48% | 1 |
British | 91.04% | 1 |
American | 88.68% | 1 |
Rate
$0.016
/min
Per-second billing. Bring your own provider key and pay your provider directly — a 5% routing fee applies (first 100 min/mo free).
Set up BYOKCost estimator
1 hour
$0.96
10 hours
$9.61
100 hours
$96.12
1,000 hours
$961.20
Billed per second of audio processed.
Model ID
google/latest-short
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 referenceSupported Languages
Supported Formats
Features
Limits
Max file size
No documented limit
Max duration
8 h