Server definition

Hash: sha256:bc41b3127556fc56dc0ad2f02346711618cea85336ba04e298ab03b5f586aadd
What it is: What a remote MCP server returned when asked what it offers: 10 tools

The blob, as servednamed by its sha256

{
  "instructions": "Speech AI API suite with 4 capabilities:\n1. **Pronunciation Assessment** — assess_pronunciation scores English pronunciation from audio at overall, sentence, word, and phoneme levels (0-100).\n2. **Speech-to-Text** — transcribe_audio converts audio to text with word-level timestamps and confidence scores.\n3. **Text-to-Speech** — synthesize_speech generates natural speech audio from text with 12 English voices, speed control, and WAV output.\n4. **Whisper STT Pro** — transcribe_audio_pro uses Whisper Large V3 Turbo for 99-language transcription with optional speaker diarization.\n\nAll tools accept base64-encoded audio. Read the resources for guides and requirements.",
  "tools": [
    {
      "description": "Assess English pronunciation quality from audio.\n\nScores pronunciation at four levels: overall, sentence, word, and phoneme.\nEach score is 0-100. Phonemes are returned in both IPA and ARPAbet notation.\nSub-300ms inference latency.\n\nArgs:\n    audio_base64: Base64-encoded audio data. Supports WAV, MP3, OGG, and WebM formats.\n    text: The reference English text that the speaker was expected to read aloud.\n    audio_format: Audio format hint — one of 'wav', 'mp3', 'ogg', 'webm'. Defaults to 'wav'.\n\nReturns:\n    dict with keys:\n        - overallScore (int 0-100): Overall pronunciation quality\n        - sentenceScore (int 0-100): Sentence-level fluency and accuracy\n        - words (list): Per-word scores, each containing:\n            - word (str): The word\n            - score (int 0-100): Word pronunciation score\n            - phonemes (list): Per-phoneme scores with IPA/ARPAbet notation\n        - decodedTranscript (str): What the model heard (ASR transcript)\n        - transcript (str): Reference text\n        - confidence (float 0-1): Scoring confidence\n        - warnings (list[str]): Quality warnings if any\n        - audioQuality (dict): Audio metrics (SNR, peak/RMS dB, etc.)",
      "inputSchema": {
        "properties": {
          "audio_base64": {
            "description": "Base64-encoded audio data. Supports WAV, MP3, OGG, and WebM formats.",
            "maxLength": 20000000,
            "type": "string"
          },
          "audio_format": {
            "default": "wav",
            "description": "Audio format hint — one of 'wav', 'mp3', 'ogg', 'webm'.",
            "type": "string"
          },
          "text": {
            "description": "The reference English text that the speaker was expected to read aloud.",
            "maxLength": 10000,
            "type": "string"
          }
        },
        "required": [
          "audio_base64",
          "text"
        ],
        "type": "object"
      },
      "name": "assess_pronunciation",
      "outputSchema": null
    },
    {
      "description": "Check if the pronunciation assessment service is healthy and ready.\n\nReturns:\n    dict with keys:\n        - status (str): 'healthy' or error state\n        - modelLoaded (bool): Whether the scoring model is loaded\n        - version (str): API version",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "check_pronunciation_service",
      "outputSchema": null
    },
    {
      "description": "Check if the speech-to-text service is healthy and ready.\n\nReturns:\n    dict with keys:\n        - status (str): 'healthy' or error state\n        - modelLoaded (bool): Whether the STT model is loaded\n        - version (str): API version",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "check_stt_service",
      "outputSchema": null
    },
    {
      "description": "Check if the text-to-speech service is healthy and ready.\n\nReturns:\n    dict with keys:\n        - status (str): 'healthy' or error state\n        - modelLoaded (bool): Whether the TTS model is loaded\n        - version (str): API version",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "check_tts_service",
      "outputSchema": null
    },
    {
      "description": "Check if the Whisper STT Pro service is healthy and ready.\n\nReturns:\n    dict with keys:\n        - status (str): 'healthy' or error state\n        - modelLoaded (bool): Whether the Whisper model is loaded\n        - diarizeLoaded (bool): Whether the diarization pipeline is loaded\n        - version (str): API version\n        - modelName (str): Whisper model name (e.g. 'large-v3-turbo')",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "check_whisper_service",
      "outputSchema": null
    },
    {
      "description": "Get the full phoneme inventory supported by the pronunciation scorer.\n\nReturns a list of all English phonemes the engine can assess, including\nARPAbet symbol, IPA equivalent, example word, and phoneme category\n(vowel, consonant, diphthong).\n\nReturns:\n    list of dicts, each with keys:\n        - arpabet (str): ARPAbet symbol (e.g. 'AA', 'TH')\n        - ipa (str): IPA notation\n        - example (str): Example word containing the phoneme\n        - category (str): vowel, consonant, or diphthong",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "get_phoneme_inventory",
      "outputSchema": null
    },
    {
      "description": "List all available text-to-speech voices with metadata.\n\nReturns:\n    dict with keys:\n        - voices (list): Available voices, each with id, name, gender, accent, grade\n        - defaultVoice (str): Default voice ID",
      "inputSchema": {
        "properties": {},
        "type": "object"
      },
      "name": "list_tts_voices",
      "outputSchema": null
    },
    {
      "description": "Generate natural speech audio from English text.\n\nProduces high-quality speech with 12 English voices.\nReturns base64-encoded WAV audio (16-bit PCM, 24kHz mono) along with metadata.\n\nAvailable voices:\n- af_heart (default), af_bella, af_nicole, af_sarah, af_sky (American female)\n- am_adam, am_michael (American male)\n- bf_emma, bf_isabella (British female)\n- bm_george, bm_lewis, bm_daniel (British male)\n\nArgs:\n    text: English text to synthesize (1-5000 characters).\n    voice: Voice ID. See list above. Defaults to 'af_heart'.\n    speed: Speed multiplier from 0.5 to 2.0 (default: 1.0).\n\nReturns:\n    dict with keys:\n        - audio_base64 (str): Base64-encoded WAV audio (16-bit PCM, 24kHz)\n        - duration_ms (str): Audio duration in milliseconds\n        - voice (str): Voice ID used\n        - text_length (str): Input text character count\n        - processing_ms (str): Synthesis time in milliseconds",
      "inputSchema": {
        "properties": {
          "speed": {
            "default": 1,
            "description": "Speech speed multiplier (0.5 = half speed, 2.0 = double).",
            "type": "number"
          },
          "text": {
            "description": "English text to convert to speech. Max 5000 characters.",
            "maxLength": 5000,
            "type": "string"
          },
          "voice": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "default": null,
            "description": "Voice ID (e.g. 'af_heart', 'am_adam'). Uses default if omitted."
          }
        },
        "required": [
          "text"
        ],
        "type": "object"
      },
      "name": "synthesize_speech",
      "outputSchema": null
    },
    {
      "description": "Transcribe audio to text with word-level timestamps.\n\nConverts spoken English audio into text with optional word-level timestamps\nand per-word confidence scores.\n\nArgs:\n    audio_base64: Base64-encoded audio data (WAV, MP3, OGG, FLAC, WebM).\n    audio_format: Audio format hint. Auto-detected from magic bytes if omitted.\n    include_timestamps: Whether to include word-level timing (default: true).\n\nReturns:\n    dict with keys:\n        - text (str): Full decoded transcript\n        - words (list): Per-word results with timestamps, each containing:\n            - word (str): The transcribed word\n            - start (float): Start time in seconds\n            - end (float): End time in seconds\n            - confidence (float 0-1): Word-level confidence\n        - audioDurationMs (int): Audio duration in milliseconds\n        - metadata (dict): Processing time, audio length, model version\n        - audioQuality (dict): Audio metrics (SNR, peak/RMS dB, etc.)",
      "inputSchema": {
        "properties": {
          "audio_base64": {
            "description": "Base64-encoded audio data. Supports WAV, MP3, OGG, FLAC, and WebM formats.",
            "maxLength": 20000000,
            "type": "string"
          },
          "audio_format": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "default": null,
            "description": "Audio format hint — 'wav', 'mp3', 'ogg', 'flac', 'webm'. Auto-detected if omitted."
          },
          "include_timestamps": {
            "default": true,
            "description": "If true, include word-level start/end times and confidence.",
            "type": "boolean"
          }
        },
        "required": [
          "audio_base64"
        ],
        "type": "object"
      },
      "name": "transcribe_audio",
      "outputSchema": null
    },
    {
      "description": "Transcribe audio with Whisper Large V3 Turbo — multilingual STT.\n\nSupports 99 languages with automatic language detection, word-level\ntimestamps, per-word confidence scores, and optional speaker diarization\n(identifies who spoke each word). Best-in-class WER (~2%).\n\nArgs:\n    audio_base64: Base64-encoded audio (WAV, MP3, OGG, FLAC, WebM).\n    language: Language code. Auto-detected if omitted. Supports 99 languages.\n    diarize: Enable speaker diarization (default: false). When true, each word\n        includes a speaker label (e.g. SPEAKER_00, SPEAKER_01).\n\nReturns:\n    dict with keys:\n        - text (str): Full decoded transcript\n        - words (list): Per-word results with timestamps, each containing:\n            - word (str), start (float), end (float), confidence (float 0-1)\n            - speaker (str|null): Speaker label when diarize=true\n        - speakers (dict|null): Speaker info with count and labels\n        - audioDurationMs (int): Audio duration in milliseconds\n        - metadata (dict): Processing time, language, languageProbability\n        - audioQuality (dict): Audio metrics (SNR, peak/RMS dB, etc.)",
      "inputSchema": {
        "properties": {
          "audio_base64": {
            "description": "Base64-encoded audio data. Supports WAV, MP3, OGG, FLAC, and WebM formats.",
            "maxLength": 20000000,
            "type": "string"
          },
          "diarize": {
            "default": false,
            "description": "Enable speaker diarization to identify who spoke each word.",
            "type": "boolean"
          },
          "language": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "null"
              }
            ],
            "default": null,
            "description": "Language code (e.g. 'en', 'es', 'zh'). Auto-detected when omitted."
          }
        },
        "required": [
          "audio_base64"
        ],
        "type": "object"
      },
      "name": "transcribe_audio_pro",
      "outputSchema": null
    }
  ]
}

Verify it yourself

curl -s https://api.teppi.xyz/v1/evidence/sha256:bc41b3127556fc56dc0ad2f02346711618cea85336ba04e298ab03b5f586aadd | sha256sum