Cantonese speech-to-text

Transcribe and translate speech across 60+ languages with the speed and accuracy real-time experiences demand.

Leading voice teams build with Deepgram

Granola logo
Twilio logo
Cloudflare logo
Vapi logo
Daily logo
Cresta logo
Decagon logo
IBM logo
Sierra logo
NICE logo

Fast and accurate Cantonese speech recognition for real-world audio

Get real-time Cantonese speech-to-text in under 300 ms while maintaining high accuracy in noisy, accented, or overlapping conversations.

Woman speaking into her phone outside an office building at dusk

Cantonese language overview

Speakers: 86 million total speakers

Regions: China (Guangdong and Guangxi), Hong Kong, Macau, Malaysia, United States, Vietnam, Singapore, Australia, United Kingdom, Canada

Dialects: Guangzhou Cantonese (standard), Hong Kong Cantonese, Macau Cantonese, Taishanese (Toisanese)

Writing system: Traditional Chinese characters with Cantonese-specific colloquial characters

Language family: Sino-Tibetan, Yue branch of Chinese languages

Cantonese is widely used across Hong Kong, southern China, and global diaspora communities, making it a key language for call center analytics, customer support AI, media captioning, healthcare transcription, multilingual voice agents, financial services, and legal documentation.

Man wearing headphones on a subway platform

Cantonese speech-to-text capabilities

  • Diarization

    Automatically detect and label who is speaking in multi-speaker Cantonese conversations.

  • Smart formatting

    Apply automatic capitalization, paragraphing, and clean transcript structure for Cantonese text.

  • Search

    Instantly find words or phrases inside long Cantonese recordings without reprocessing audio.

  • Utterances

    Segment streaming Cantonese audio into real-time sentence-level units for voice agents.

  • Punctuation

    Add accurate punctuation and capitalization to Cantonese transcripts for easy reading.

  • Redaction

    Automatically remove sensitive data like credit cards, phone numbers, and PII from Cantonese transcripts.

Keyword boosting for Cantonese

Amber bakelite relief carved with letters

Voice AI that performs in the real world

Low latency. Effortless scale. Production-grade reliability. Everything you need to build and run real-time voice experiences.

Bakelite radio player on a blue halftone tile

Domain-aware speech

Understand specialized terminology, names, accents, and industry-specific language with production-grade accuracy.

Copper megaphone on an orange halftone tile

Expressive voices

Give every interaction a natural, consistent voice that adapts its tone when the conversation changes.

Glass globe on a violet halftone tile

Context-aware delivery

Carry meaning, tone, and conversational state across turns so agents listen and respond more naturally.

Scale beyond Cantonese with one API

Start with Cantonese speech-to-text, then expand to 45+ languages using the same API, models, and tooling.

  • English

    English

  • Spanish

    Español

  • German

    Deutsch

  • Japanese

    日本語

  • French

    Français

  • Portuguese

    Português

  • Mandarin

    中文(普通话)

  • Korean

    한국어

  • Hindi

    हिन्दी

  • Arabic

    العربية

  • Turkish

    Türkçe

  • Italian

    Italiano

FAQ

Ready to build with Cantonese speech-to-text?

Start transcribing Cantonese audio with Deepgram's speech-to-text API. It is fast, accurate, and built for real-time applications.

See Deepgram in action

Loading form