Bug LiveOptions Configuration for Cantonese (zh-HK) is Detecting Chinese rather than Cantonese (zh-HK)

[mr-haseeb]

Hi Deepgram Team,
The deepgram product you are using : Speech to Text
I hope you're doing well.

We’re currently working on real-time transcription for Cantonese (zh-HK), and had a few questions regarding the LiveOptions configuration. Below is a snippet of our current setup:

# Deepgram client config
config = DeepgramClientOptions(
    # options={"keepalive": "true"},
    # verbose=verboselogs.DEBUG,
)

deepgram = DeepgramClient(
    api_key=get_secret("DEEPGRAM_API_KEY"),
    config=config,
)

# Create websocket connection
client.ws= deepgram.listen.websocket.v("1")

live_options = LiveOptions(
    model="nova-2",
    language="zh-HK",
    smart_format=True,
    encoding="linear16",
    channels=1,
    sample_rate=16000,
    interim_results=True,
    utterance_end_ms="4600",
    vad_events=True,
    endpointing="4600",
)

However, the returned transcription appears to be detecting the language as general Chinese rather than Cantonese (zh-HK). We're wondering if there’s any adjustment needed in the configuration or if there are known limitations for Cantonese recognition with nova-2.

Any guidance or confirmation would be greatly appreciated!

Thanks,
Haseeb

[deepgram-community[bot]]

Thanks for asking your question. Please be sure to reply with as much detail as possible so the community can assist you efficiently.
_{Consider joining our Discord community for more opportunity to engage with your fellow Deepgram users. You can earn points which can be redeemed for cool stuff by being active in our communities!}

[deepgram-community[bot]]

Hey there! It looks like you haven't connected your GitHub account to your Deepgram account. You can do this at https://community.deepgram.com - being verified through this process will allow our team to help you in a much more streamlined fashion.

[deepgram-community[bot]]

It looks like we're missing some important information to help debug your issue. Would you mind providing us with the following details in a reply?

• The deepgram product you are using (e.g Speech to Text, Agent API)
• A request ID that triggered your error or issue.

[jkroll-deepgram]

Hi @mr-haseeb, can you share what discrepancy you're observing? Because you are setting the language as zh-HK, Deepgram will transcribe the audio as Cantonese, rather than doing any language detection.

If you can share any Deepgram request IDs associated with your testing, that would be helpful for us to troubleshoot as well.