Optimized for low-bandwidth single speaker voicemail
1 languages · benchmarked
Nova-2 Voicemail is a speech-to-text model from Deepgram. In our standardized benchmarks it reaches 17.51% word error rate, ranking #22 of 35 models tested. It supports 1 language and runs at $0.006/min — mid-range on price.
Its nearest benchmarked alternative is Soniox STT (Async). Best suited for live captioning and streaming transcription.
Overall Score
76.7
#22 of 35
Word Error Rate
17.51%
Character Error Rate
12.36%
Match Error Rate
17.10%
Word Info Lost
22.71%
Avg Latency
954ms
Benchmarks Run
25
last 30 runs
Word error rate
17.51%
↓ 2.7% · 30d
5 runs ago
latest
Avg latency
954ms
↓ 22.3% · 30d
5 runs ago
latest
overall WER · 35 models
Nova-2 Voicemail
field · 34 models
lower-left is better
8 categories
WER
lower is better
Legal
4.2%
Technical
5.5%
Medical
7.3%
Conversational
10.8%
Finance
11.3%
General
11.6%
Noisy Environment
31.3%
Code-Switching
65.3%
p50 → p99 · 5 runs
Latency distribution
lower is better
1.6s
p50
1.7s
p90
1.7s
p95
1.7s
p99
5 accents
Realtime Score
77.2
TTFW (P50)
1014 ms
Flicker
10.1%
Cadence
0.8/s
RTF
1.02×
Word Error Rate
19.27%
Character Error Rate
12.34%
Benchmarks Run
25
Rate
$0.0058
/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
$0.35
10 hours
$3.49
100 hours
$34.92
1,000 hours
$349.20
Billed per second of audio processed.
Model ID
deepgram/nova-2-voicemail
Authenticate every request with your secret API key as a Bearer token. Issue a key from your dashboard.
This model also supports realtime streaming over WebSocket.
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.
custom_words
string[]
Names, jargon and product terms to bias the model toward. Up to 1000 entries.
vocabulary_list_id
uuid
A vocabulary list saved in your workspace. Merged with custom_words when both are sent.
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
2 GB
Max duration
No documented limit