Thai THAI-SER

by VISTEC / NECTEC

36h Thai emotional-speech corpus — the open Thai SER + ASR baseline.

TL;DR

36h Thai emotional-speech corpus — the open Thai SER + ASR baseline.

Best for thai emotion recognition + Thai ASR research. Pricing: free.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free
Platforms
HuggingFace

What it is

THAI-SER is a 36-hour Thai SER corpus with 200 actors across 5 emotion classes. License: CC BY-SA 4.0.

Best for: Thai emotion recognition + Thai ASR research.
Watch out for: CC BY-SA 4.0 · 200 actors · 5 emotion classes. Cite: Pongkittiphan et al., O-COCOSDA 2021.

Install / use

from datasets import load_dataset; ds = load_dataset('vistec-AI/thai-ser')

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported1
HIPAA eligibleNo

Thai THAI-SER vs Whipscribe

FeatureThai THAI-SERWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsHuggingFaceWeb, API, MCP

Alternatives to Thai THAI-SER

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

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.