Teochew Whisper

Teochew Whisper

A Gradio demo of a Whisper speech model fine-tuned for Teochew, for people who want to try speech recognition for this Chinese regional language in a browser or on their own computer.

Made by panlr on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: Teochew Whisper is a speech recognition demo for Teochew, a Chinese regional language. Going by the model name panlr/whisper-finetune-teochew, it runs a Whisper model that was fine-tuned for Teochew speech. The repository also holds a sample audio file called example.wav.

Who it suits: It suits people who work with Teochew speech and want to see what automatic transcription can do with it.

Title: Who might use this demo
Industry | Who uses it | How they use it | Example
Linguistics | Researchers | Test automatic transcription on a regional language | Play a short Teochew recording through the demo
Speech technology | Developers | See how a fine-tuned Whisper model handles a small language | Compare its output with a transcript you trust
Language learning | Learners and heritage speakers | Check a spoken phrase against written text | Record a short phrase and read what comes back

If you need another language or dialect, you can skip it.

Where to use it: You can use it online or run a copy yourself.

• Web browser: the Space's page on Hugging Face, with nothing to install
• Windows, macOS or Linux: a local copy that runs with Python
• Hugging Face Spaces: where the hosted demo runs, on the cpu-basic hardware tier

Title: What the demo is good for
Use case | Who | What they get back
Transcribe Teochew speech | Researchers | Text produced from a spoken recording
Try the sample clip | Anyone curious | The result for the bundled example.wav file
Test a fine-tuned model | Developers | A quick look at how it performs on audio they choose

How to set it up: The hosted demo needs nothing. A local copy needs Python and pip.

• Python, with pip
• The Python packages that requirements.txt lists

1. Install the huggingface_hub package.
2. Download the Space's files into a folder.
3. Move into that folder.
4. Install Gradio 5.21.0.
5. Install the packages from requirements.txt.
6. Start the app with Python.

pip install -U huggingface_hub
hf download panlr/teochew_whisper --repo-type space --local-dir teochew_whisper
cd teochew_whisper
pip install gradio==5.21.0
pip install -r requirements.txt
python app.py

How to use it: Open the Space online, or start your copy with the command below. Then use the controls on the page to supply a Teochew recording and read the text it returns.

python app.py

Title: The files you will meet
Tool | What it does
app.py | Starts the Gradio page when you run it with Python
example.wav | A sample audio file you can use to try the demo
requirements.txt | Lists the Python packages the app needs

Example: You open the Space and give it a short Teochew recording, such as the bundled example.wav. The page returns the words it heard as written text.

For newcomers and experts: If you are new to this, try the sample clip first, then a short recording of your own. Short, clear audio gives the cleanest test.

If you know the field, the demo wraps a community fine-tune of Whisper for a language that general speech models cover thinly. The front end is Gradio 5.21.0, and the hosted copy runs on a CPU tier. The notes give no accuracy figures or training data details.

Open the Space or run app.py
Supply a Teochew recording
The fine-tuned Whisper model listens
You read the text it returns

Good to know: The Space's notes are brief, so open the page to see exactly what it returns.

• Hosting: the hosted copy runs on cpu-basic hardware
• Notes: the README only points to Hugging Face's configuration reference
• License: cc-by-4
• Cost to run: none named
• Data it reaches: the audio you supply, plus the Python packages pip installs from requirements.txt
• Not for: languages other than Teochew

On Hugging Face, 13 November 2026

• Likes: 7
• Built with: gradio
• License: cc-by-4.0

Links

• Space: https://huggingface.co/spaces/panlr/teochew_whisper
• Files: https://huggingface.co/spaces/panlr/teochew_whisper/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.