Naver Clova Speech

by Naver Cloud Platform

Naver Cloud's Korean-first ASR with batch + real-time and speaker diarization.

TL;DR

Naver Cloud's Korean-first ASR with batch + real-time and speaker diarization.

Best for korean-language meeting, broadcast, and contact-center workflows. Pricing: tiered KRW-per-second.

Category
Transcription APIs
License
—
Stars
—
Last push
—
Pricing
tiered KRW-per-second
Platforms
API

What it is

Naver Cloud's Clova Speech family includes short-utterance recognition (CSR) for under-60-second clips and the longer-format Clova Speech endpoint for full meetings and broadcasts with speaker diarization, word timestamps, and segmentation. Korean is the strongest language; English, Japanese and Mandarin are supported on specific endpoints. Pricing is KRW-denominated with tiered per-second rates. Strong fit for Korean SaaS and broadcast pipelines. Best fit: korean-language meeting, broadcast, and contact-center workflows. Caveats: korean-first; english and japanese supported but quality varies; docs primarily in korean. Pricing as listed: tiered KRW-per-second. Feature flags from vendor docs: speaker diarization, word-level timestamps, streaming. Directory tags: commercial-api, regional-asia. Last vendor-page check: 2026-05-12.

Best for: Korean-language meeting, broadcast, and contact-center workflows.
Watch out for: Korean-first; English and Japanese supported but quality varies; docs primarily in Korean.

Install / use

Naver Cloud: Clova Speech Recognition (CSR) / Long Sentence (CLOVA Speech)

Features

Speaker diarizationYes
Word-level timestampsYes
Streaming / real-timeYes
Languages supportedNone
HIPAA eligibleNo

Naver Clova Speech vs Whipscribe

FeatureNaver Clova SpeechWhipscribe
CategoryTranscription APIsTranscription APIs
Pricingtiered KRW-per-secondfree beta
Speaker diarizationYesYes
Word timestampsYesYes
StreamingYesNo
Languages—99
PlatformsAPIWeb, API, MCP

Alternatives to Naver Clova Speech

OpenAI Whisper API
OpenAI

Hosted Whisper large-v3 from OpenAI — $0.006 per minute.

$0.006/min
AssemblyAI
AssemblyAI

Universal-2 model + diarization, PII redaction, topic detection, summarization.

from $0.37/hr
Deepgram
Deepgram

Nova-2 model, excellent streaming, strong at conversational audio.

from $0.0043/min

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.