$ loading_
Transcribe, translate, and synthesize English-Chinese speech from WAV audio files.
The materials indicate the tool mainly processes audio locally and executes local transcription/TTS, with no declared required secrets or fixed remote endpoints. No concrete high-risk red flags are evident, but as a third-party open-source MCP that executes locally, it still warrants normal caution around runtime behavior and dependencies.
The materials list no required secrets or environment variables, and there is no explicit request for API keys, tokens, or local sensitive credentials. Although the description mentions translation 'via Claude API,' the tool itself does not specify any secret configuration, so no clear credential exposure point is shown in the provided materials.
The description says translation occurs 'via Claude API,' which implies transcribed text or related content may be sent to an external AI service. No specific host is declared and the system check lists no remote endpoint, but normal outbound data flow is still possible and the actual destinations and payloads should be verified.
The system marks this tool as executes-code, and the description indicates it runs local Whisper and Piper TTS processing. This is a typical local execution capability for an MCP/tool, and no unusual system privileges or functionally unrelated dangerous execution behavior are evident from the materials.
The tool requires a WAV file path, so it will at least read user-specified local audio files and may produce intermediate transcripts or audio output. No data access beyond what is needed for speech translation is apparent, but it still has ordinary local file access capability.
Positive signals include being open source under the MIT license, which makes code review possible. However, it comes from a third-party registry, has 0 stars, and has unknown maintenance status, so its trust and maturity are limited and dependency/maintenance risks should be independently checked.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "live-translate-mcp" yet — see the docs or source repo.
Please process this WAV file path: /audio/team_meeting.wav. First transcribe the original speech, then translate English to Chinese or Chinese to English, and finally generate synthesized speech output. Provide both the text and a description of the audio result.
Returns the transcript, the translated text, and a description of the synthesized speech output.
Please read /audio/practice.wav, translate the spoken Chinese into natural English, and generate English speech; if the original audio is in English, translate it into Chinese and generate Chinese speech. Also point out any unnatural or unclear expressions.
Returns the transcript, polished translation, notes on unclear phrasing, and the target-language speech output.
Process the WAV file /audio/interview.wav, detect whether the interview is in English or Chinese, translate it into the other language, and output structured text suitable for interview notes: show the original and translated content in segmented blocks.
Returns segmented original and translated text, ready for interview notes or content organization.
Capture Vietnamese or English speech and paste cleaned text into Claude Code.
Convert text to speech quickly and browse available voices and models.
Convert speech and text with OpenAI-compatible transcription and synthesis APIs.
Convert text to speech with inline playback and history in MCP hosts.
Download YouTube videos and transcribe speech locally with Whisper.
Transcribe audio into multilingual text with optional GPT-based post-processing.