Best Cantonese Speech-to-Text Tools in 2026: We Tested 2, and This One Is Most Reliable for Students

Last semester, I took a general education course on Hong Kong culture. The professor lectured entirely in Cantonese, and the slides only had English outlines. I tried taking notes by hand, but when I reviewed them before the midterm, I couldn't even understand my own Cantonese romanization. So I started looking for speech-to-text tools, trying everything from free to paid options, and finally settled on Tinrec and Otter. This article shares my hands-on experience from the semester, comparing them across four dimensions: Cantonese accuracy, free tier, organization features, and export flexibility, to help you avoid wasting time on the wrong tool.

Tinrec vs. Otter: Quick Overview

Tinrec is an AI-powered audio and video organization tool that turns recordings, videos, or online clips into searchable, summarizable, and exportable content. Otter is a well-established AI meeting notes tool, excelling in English online meetings and team collaboration. Both can do speech-to-text, but for Cantonese speakers and student use cases, the differences are significant.

Feature	Tinrec	Otter
Core Positioning	Multi-source audio/video organization workspace	English meeting notes and team collaboration
Free Tier	Monthly base transcription quota; core features available	300 minutes of transcription per month
Input Sources	Live recording, audio/video files, online video links	Meeting platforms, file import
Cantonese Support	Optimized for Chinese content; stable Cantonese recognition	Primarily English; Cantonese often mis-transcribed as Mandarin
Organization Features	AI summaries, chapters, to-dos, Q&A, mind maps	Meeting summaries, AI chat, action items

Below, I'll compare the four key dimensions in detail, each based on my actual tests using the same lecture recordings or discussion audio.

Dimension 1: Cantonese Recognition Accuracy

I initially used Otter for English group discussions, and its accuracy was decent. But when I switched to Cantonese lecture recordings, problems emerged—it often transcribed "我哋" as "我們," "食飯" as "吃飯," and even converted entire Cantonese colloquial sentences into written Mandarin, which wasn't the verbatim transcript I needed. Later, I tested the same recording with a classmate, and Tinrec clearly understood Cantonese better: it preserved colloquial terms like "係," "唔該," and "好叻," and handled mixed Cantonese-English without glitches.

→ This dimension: Tinrec wins clearly. If you need Cantonese speech-to-text, Otter really isn't up to the task.

Dimension 2: Free Tier and Student Budget

Otter offers 300 free minutes per month, which sounds generous, but if you're mainly transcribing Cantonese, those 300 minutes are almost wasted—the output isn't usable without heavy editing. Tinrec's free tier doesn't publicly specify minutes, but with my usage of two lectures per week plus occasional group discussions, the free quota was sufficient, and I didn't feel pressured to upgrade. If you do pay, the weekly or Pro monthly card costs about the same as a bubble tea or two late-night snacks—affordable for students.

→ This dimension: Tinrec wins. The "usability" of the free tier matters more than raw minutes.

Dimension 3: Post-Transcription Organization Features

Otter automatically generates summaries and action items for meetings and offers AI chat, but in my experience, its summaries are English-centric and often miss the key points in Chinese content. Tinrec's organization features are far more comprehensive: besides AI summaries, it auto-chapters, extracts to-dos, and can even turn long lecture recordings into mind maps. My favorite is the "AI Q&A"—before finals, I'd ask the recording, "What are the key exam points the professor mentioned?" It would find answers from the recording, saving me hours of re-listening.

→ This dimension: Tinrec wins. Transcription is just the starting point; post-processing is where the real time savings happen.

Dimension 4: Export Flexibility and Integration

Otter allows exporting transcripts and summaries, with support for Google Docs, but format options are limited. Tinrec supports TXT, DOCX, PDF, SRT, VTT, Markdown, and more, and can sync directly to Notion, OneNote, and Dropbox. I personally convert lecture recordings to Markdown and put them in my Notion notes, linking to my review system—something Otter can't do as smoothly.

→ This dimension: Tinrec wins. If you have a fixed note-taking or knowledge management system, Tinrec's integration is stronger.

What Unique Advantages Does Each Offer?

Tinrec's Unique Advantages

1. Online Video Transcription: Paste a YouTube or Bilibili link to transcribe; I used it to organize Hong Kong news clips for reports, super convenient.
2. AI Q&A and Agent Post-Processing: Not just asking questions, but also generating meeting minutes, report tables, and even completing assignments.
3. Optimized for Chinese Content: Handles Cantonese, Mandarin, and mixed Chinese-English well, unlike Otter which is English-oriented.
4. Multi-format Export and Third-party Sync: Covers almost all the tools I use, ensuring a seamless workflow.

Otter's Unique Advantages

1. English Meeting Collaboration: Real-time transcription and team collaboration for English online meetings are very mature; if you frequently have English meetings, it's still the top choice.
2. Generous Free Minutes: 300 minutes is plenty for light English users, provided you don't need Cantonese.

Summary: Which One Should You Choose?

• If you need Cantonese speech-to-text, Chinese course notes, or interview transcription → Tinrec (wins on Cantonese accuracy, organization, and export flexibility)
• If you frequently have English online meetings and need team collaboration → Otter is more suitable
• If you're a student like me, on a budget, and want one tool for recording, transcription, and organization → Start with Tinrec's free version; it's genuinely sufficient

In a nutshell: For Cantonese speakers and student use cases, Tinrec is the more complete choice. I recommend recording a sample of your own lecture with the free version to test Cantonese recognition and AI organization before deciding on an upgrade.

References

• Transcribe Cantonese Audio to Text Free in Seconds
• AI Speech-to-Text: Fast Audio Transcription, Subtitles & Captions - Free Trial - cSubtitle
• AI Cantonese Speech-to-Text: Professional Cantonese Recognition
• Online AI Speech, Video, Audio Auto-Transcription - Free Trial - cSubtitle
• Free + Cantonese OK: How to Use Google AI Studio to Transcribe Your Recordings...