Skip to content

Support additional STT providers (Tencent Cloud / Aliyun / iFlytek …) #173

Description

@apple-art

Is your feature request related to a problem?

Glass currently relies on Gemini Live STT or local Whisper.
Many users would benefit from additional, inexpensive speech-to-text APIs that already provide high-quality realtime transcription.

Describe the solution you’d like

• Add adapter(s) for at least one of the following vendors:
– Tencent Cloud Realtime ASR (wss://asr.cloud.tencent.com/...)
– Aliyun Realtime ASR (wss://nls-gateway.cn-shanghai.aliyuncs.com/...)
– iFlytek RT-ASR

• Keep the public STT interface unchanged (createSTT({ language, callbacks … })), with provider-specific logic placed in new modules.

• Credentials could be supplied via environment variables or packed into a single apiKey string to avoid UI changes.

Describe alternatives you’ve considered

Local Whisper works but is CPU/GPU-heavy and not as cheap as cloud minutes; Gemini Live is great but costs more and offers fewer regional pricing options.

Additional context

– Tencent Cloud docs: https://cloud.tencent.com/document/product/1093/48982
– Aliyun docs: https://help.aliyun.com/document_detail/179272.html
– iFlytek docs: https://www.xfyun.cn/doc/asr/rtasr/API.html

Happy to help with a PR or testing if this aligns with the roadmap.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions