Turn recordings into transcripts and summaries in minutes
Upload audio or video for multilingual transcription, AI notes, and action items
5 Cantonese Speech-to-Text Tools Tested for 2026: Free Tiers and Cantonese Accuracy Compared
"Free" can mean three very different things when it comes to Cantonese speech-to-text.
Some tools give you a fixed number of minutes per month, some let you try only the first 15 minutes of each file, and some are built into your phone and cost nothing but only support real-time dictation—you can't upload audio files.
So the first question isn't "which one is most accurate?" but "how long do I record each month, and do I need to turn the transcript into meeting notes or subtitles afterward?"
This article lines up 6 common solutions and looks at them from four angles: free quota, Cantonese recognition, mixed Chinese-English, and spoken-to-written conversion. The quotas and prices below are based on each tool's official announcements; we recommend testing with your own recordings before committing.
Why do you need a tool specifically for Cantonese transcription?
The challenge with Cantonese isn't "can it hear the words?" but "can you use the output?"
Manually typing subtitles for a 20-minute video often takes an hour, and colloquial phrases like "係咪" (is it?), "唔該" (please/thanks), and "搞掂" (done) still need to be rewritten entirely when they go into formal records.
Plus, Hong Kong meetings often mix Chinese and English, and tools optimized only for English or Mandarin often produce gibberish that needs retyping.
This article covers 6 solutions for this scenario, from free trials to team-ready options.
Solution 1 | Tinrec — The most complete AI recording assistant
Tinrec is an AI meeting notes and collaboration tool for individuals and teams. It doesn't just convert audio to text—it organizes meetings into searchable, queryable, and exportable data.
Record online meetings without inviting a bot. The desktop app captures system audio directly, handling meetings played on your computer via Zoom, Google Meet, Microsoft Teams, and Webex—no extra bot needs to join the meeting, and a transcript is generated as you record.
The real value comes after transcription. The tool automatically generates summaries, chapters, key points, and action items, and you can ask questions about the meeting content directly, such as "What were the conclusions of this meeting?"—no need to scroll through a 30-minute transcript. Outputs can be exported in common formats and brought into existing workflows like Notion or Google Docs.
Multilingual and mixed Chinese-English. Supports transcription for multilingual meetings including Chinese, Cantonese, and English, with real-time translation and viewing in original, translated, or bilingual modes. Team hotwords let you add company-specific terminology to reduce misheard proper nouns.
Free and paid plans. The free version offers a basic transcription quota (about 100 minutes per month; check the official website for current details). Paid options include a weekly pass, Pro monthly, and Pro annual, depending on usage frequency. The team version is a separate team space: a 7-day trial for eligible teams with 1 free seat and 300 shared import minutes; monthly USD 29.80 per paid seat per month, annual USD 199 per paid seat per year (about USD 16.58 per month), with each paid seat providing 2,000 shared import minutes per month. Under an active team plan and active seat, real-time recording currently does not deduct minutes.
Who it's for: People who record several meetings a month and want meeting notes and action items right after recording, or who need to store meeting data in a team space.
Keep in mind: AI summaries and transcriptions can still miss details; for important content, cross-check with the transcript and recording playback. Actual recognition performance depends on recording quality, accent, background noise, and overlapping speakers.
Solution 2 | Subanana — The Cantonese model specialist
In a nutshell: an online tool that turns Cantonese conversations into speaker-labeled, punctuated transcripts.
Upload an audio or video file, or paste a YouTube link, and the AI outputs a transcript with speakers separated and punctuation added. It handles mixed Chinese-English, lets you choose between colloquial and written styles, and outputs in both Traditional and Simplified Chinese. It supports export to SRT, VTT, TXT, DOCX, XLSX, and Markdown, and offers a timeline with waveforms and custom vocabulary lists.
The free option is "preview the first 15 minutes of each file for free," not a monthly minute allowance.
Who it's for: Those who mainly work with Cantonese content and need subtitle files; good flexibility for short clips or organizing YouTube videos.
Stop organizing recordings by hand
Upload audio or video and automatically get a transcript, summary, and action items
Solution 3 | Notta — A veteran choice for multilingual transcription
In a nutshell: a multilingual tool covering transcription across devices, meeting notes, and file import.
The free version offers 120 minutes per month, but each session is limited to 3 minutes. Pro annual billing is about USD 8.17 per month for 1,800 minutes per month; Business annual billing is about USD 16.67 per seat per month, with role permissions and usage reports.
Multilingual support is its strength, covering both meetings and uploaded file transcription.
Who it's for: Those who need multilingual transcription and subtitles. Note the 3-minute per-session limit, which isn't friendly for hour-long meeting recordings.
Solution 4 | TurboScribe — For long files and batch processing
In a nutshell: a file-transcription-focused tool built for long documents and batch uploads.
The free version allows 3 files per day, up to 30 minutes per file; Unlimited annual billing is USD 120 (about USD 10 per month). Paid versions support files up to 10 hours or 5 GB, up to 50 files per upload, plus subtitles and translation.
Its strength is "volume," not real-time meeting processing. For Cantonese performance, we recommend testing with your own recordings before deciding.
Who it's for: Those with a batch of recordings to process all at once.
Solution 5 | Google Voice Typing — Completely free real-time dictation
In a nutshell: a zero-cost dictation feature built into Android and iPhone.
Completely free, no registration or installation required. In a quiet environment, it performs reasonably well for short dictation.
But its limitations are clear: it doesn't support uploading audio files, only real-time dictation. With long sentences, colloquial Cantonese, or background noise, accuracy drops—common issues include "唔該" being heard as "唔乖" or a whole sentence missing a "嘅" character.
Who it's for: Those who just want to quickly note a sentence or two without installing any tool.
Solution 6 | Otter.ai — Best for English meetings
In a nutshell: a mature English AI meeting assistant.
The free Basic plan offers 300 minutes per month—a generous quota. Pro annual billing is about USD 8.49 per user per month, including 1,200 minutes of in-app recording and 10 file imports per month. Automatic summaries and speaker identification are quite mature.
However, it doesn't support Cantonese, so if your meetings are mainly in English, it's more cost-effective than the options above; for Cantonese recordings, you'll need another tool.
Who it's for: Those whose meetings are mainly in English and who need mature summary features.
With so many options, how should you choose?
• Want transcription, summaries, action items, and AI Q&A all in one → Tinrec (few solutions offer all of these) • Need to handle meetings mixing Cantonese, Chinese, and English → Tinrec (multilingual meeting transcription plus real-time translation) • Want to switch between mobile, desktop, and web → Tinrec (multi-platform, desktop can record online meetings) • Need to share meeting data with a team and manage member seats and usage → Tinrec Team (separate management of roles, seats, usage analytics, and audit logs) • Only need Cantonese subtitle files, mainly short clips → Subanana (free preview of first 15 minutes per file) • Have a batch of long recordings to process → TurboScribe (strong for long files and batch uploads) • Mainly English meetings → Otter.ai (free 300 minutes per month) • Zero budget, only need real-time short dictation → Google Voice Typing
4 things to know before using free Cantonese speech-to-text
1. "Free" is calculated in completely different ways. Some give a fixed number of minutes per month (Tinrec, Notta, Otter), some offer a 15-minute preview per file (Subanana), some have a 3-minute per-session limit (Notta free), and some are completely unlimited but feature-limited (Google Voice Typing). Know which type it is first to judge whether it's enough.
2. Don't judge accuracy by marketing numbers alone. Many tools claim high accuracy, but that's usually measured in a quiet studio. Real meetings have background noise, overlapping speakers, and mixed Chinese-English, so performance will be lower. The most practical approach is to test with a recording of your own.
3. Spoken-to-written conversion determines how much editing you'll do afterward. Cantonese colloquialisms like "係咪," "唔該," and "搞掂" need to be written as "是否," "謝謝," and "完成" in formal records. If a tool only types what it hears, the editing workload will be huge.
4. Free quotas run out. Estimate how long you record each month before deciding whether to upgrade. Tools like Tinrec that give a fixed monthly quota require waiting for the next cycle when you run out—hitting the limit at the end of the month can be frustrating.
Conclusion
Free Cantonese speech-to-text tools do exist, but each "free" has a different definition.
If you only occasionally organize a short recording, Subanana's free preview or Google Voice Typing is enough.
But if you have meetings and interviews to process every month and want summaries and action items right after transcription, a tool like Tinrec that extends from "transcription" to "organization and Q&A" will save you more time than a simple transcript tool.
We recommend starting with the free version, testing with your own recordings, and then deciding whether to upgrade.
References
- Free Cantonese Speech-to-Text App 2026 | Subanana
- 2026 Cantonese Speech-to-Text Free Solutions: 4 Tools at a Glance - Tinrec Blog
- [Free Tool] Cantonese Speech-to-Text! Just Upload a File or YouTube Link. Up to 96% Accuracy - YouTuber AI | Spend time on creativity, leave the AI to do the hustle work.
- 2026 Cantonese Speech-to-Text Free Solutions: 2 Tools at a Glance - Tinrec Blog
- 7 Recommended Cantonese/Chinese Speech-to-Text Tools for 2026: Free Online Solutions for Meeting Transcript Pain Points - Tinrec Blog
Turn every recording into actionable outcomes
Get 60 free transcription minutes when you sign in. No credit card required.
Related Reading
You might also like
2026 iPhone Speech-to-Text Buying Guide: 3 Methods Tested and Recommended
Starting with the transcription limitations of iPhone Voice Memos, this guide outlines practical workflows for converting iPhone recordings into text files in 2026, and uses Tinrec to demonstrate how to build a sustainable meeting notes workflow from transcription and summarization to to-dos and team sharing.
4 Cantonese Speech-to-Text Tools Tested: Which Handles Mixed Chinese-English and Spoken-to-Written Conversion Best?
The real challenge of Cantonese speech-to-text isn't typing—it's converting spoken language to written form, handling mixed Chinese-English, and post-meeting organization all at once. Based on real workflows, this article breaks down 5 key purchasing criteria, compares Tinrec, Subanana, CantoSub, and Otter.ai for different scenarios, and lists 4 common pitfalls plus a 6-step starter checklist.
5 Google Speech-to-Text Tools Tested for 2026: Is the Free Version Enough?
Google Docs' built-in voice typing is free and handy, but it falls short for meeting notes and existing recordings. This article walks through a real workflow and compares five approaches—Google Docs, Google Cloud Speech-to-Text, Tinrec, Notta, and Otter.ai—breaking down input sources, post-meeting organization, team collaboration, and pricing to help you decide if the free version is enough and when to switch tools.
4 Google Speech-to-Text Apps Tested: Live Transcribe, Pixel Recorder, and Tinrec Compared (2026)
Most people searching for a "Google speech-to-text app" are really trying to solve the problem of never finishing meeting notes. This guide breaks down Google Live Transcribe, the Pixel Recorder app, Cloud Speech-to-Text API, and Tinrec based on real-world use cases, with buying criteria, pitfalls to avoid, and a post-meeting follow-up template to help you pick the one that actually saves you time.
2026 Google Speech-to-Text Guide: 4 Tools Tested and Recommended
A breakdown of three common ways to convert audio to text with Google, plus a comparison of how Tinrec differs for meeting recording, post-meeting organization, and team collaboration, helping you choose the right tool for your use case.
6 Google Meet Speech-to-Text Tools Tested and Compared for 2026: How to Choose Between Free and Paid
How do you transcribe Google Meet calls? This guide covers six practical methods, from Google Meet's built-in transcription and Google Docs' free dictation to AI tools like Tinrec, Notta, TurboScribe, and Otter. We compare free tiers, use cases, and pricing to help you choose the right solution for your meeting needs.
5 Free Google Speech-to-Text Options in 2026: A Complete Guide
Want to turn recordings into transcripts without spending money? This guide covers 5 free speech-to-text options in the Google ecosystem in 2026, from Google Docs voice typing and Android Live Transcribe to the Gemini app, Gemini Notebook, and Google AI Studio. We explain the free limits, length caps, and best use cases for each, plus a practical method for better meeting notes.
2026 Google AI Studio Speech-to-Text Buyer's Guide: 4 Tools Tested and How to Choose
Can Google AI Studio really turn recordings into transcripts for free? This guide tackles common misconceptions and compares four speech-to-text approaches—Google AI Studio, Tinrec, Notta, and Otter.ai—covering audio length limits, post-transcription processing, and team collaboration differences to help you pick the right tool for your workflow.
2026 Comparison of 4 Cantonese Speech-to-Text AIs: Which Saves the Most Time on Cantonese Transcripts?
Starting from the decision focus of 'what will this recording ultimately become,' this guide walks you through the complete Cantonese speech-to-text workflow: defining output format, ensuring recording quality, AI transcription, hotword proofreading, post-meeting organization, and export. It uses Tinrec's five-step process and explains the applicable boundaries of Cantonese-specific tools.