{ "title": "Speaker diarization", "human_url": "https://happytails.ai/en/speech-tools/speaker-diarization/", "agent_url": "https://happytails.ai/agents/en/speech-tools/speaker-diarization/", "language": "en", "markdown_url": "https://happytails.ai/en/speech-tools/speaker-diarization.md", "content": "Speaker diarization\nGroup non-overlapping voices using local ECAPA speaker embeddings.\nHow to use Speaker diarization\nChoose a file in WAV, MP3, M4A, FLAC, MP4, MOV or WEBM.\nReview the settings, then choose Run Speaker diarization.\nReview the result, then use Copy result or a download link when available.\nWhat to expect\nGroup non-overlapping voices using local ECAPA speaker embeddings. Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.", "content_format": "text/plain", "visibility": "public", "links": [], "tool": { "id": "speaker-diarization", "category": "speech-tools", "built": true, "requirements": "Label speaker turns as Speaker 1 and Speaker 2; let users merge and rename speakers; review overlap; export timestamped JSON. Does not identify people by voice.", "implementation": { "fields": [ { "key": "speakers", "label": "Number of speakers (0 to estimate)", "type": "number", "value": 0, "min": 0, "max": 6, "step": 1 } ], "files": true, "accept": ".wav,.mp3,.m4a,.flac,.mp4,.mov,.webm", "multiple": false, "input": false, "processing": "Server required", "native": true, "ai": false, "maxFileBytes": 20971520, "note": "Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.", "extra": true, "experience": { "automatic": false, "single": false, "label": "Your text", "placeholder": "Type or paste your input…", "help": "Up to 1,000,000 characters. Review the settings before running.", "action": null }, "mode": "explicit", "interactive": false }, "execution": "POST /api/v1/tools/speaker-diarization/run", "api": { "id": "speaker-diarization", "name": "Speaker diarization", "category": "speech-tools", "description": "Group non-overlapping voices using local ECAPA speaker embeddings.", "notes": "Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.", "human_path": "/en/speech-tools/speaker-diarization/", "method": "POST", "endpoint": "/api/v1/tools/speaker-diarization/run", "schema_url": "/api/v1/tools/speaker-diarization", "input_schema": { "type": "object", "additionalProperties": false, "properties": { "input": { "type": "string", "maxLength": 1000000, "default": "", "description": "Plain text input. File tools use files instead unless otherwise documented." }, "options": { "type": "object", "additionalProperties": false, "properties": { "speakers": { "type": "number", "description": "Number of speakers (0 to estimate)", "default": 0.0, "minimum": 0, "maximum": 6, "multipleOf": 1 } } }, "files": { "type": "array", "maxItems": 1, "items": { "type": "object", "required": [ "name", "base64" ], "additionalProperties": false, "properties": { "name": { "type": "string", "maxLength": 200, "description": "Filename only, no path." }, "mime": { "type": "string", "maxLength": 150 }, "bytes": { "type": "integer", "minimum": 0, "description": "Optional decoded byte count; must match content if supplied." }, "base64": { "type": "string", "contentEncoding": "base64", "description": "File bytes as padded base64. Use the tool-specific limits.files_bytes value for the total decoded input size." } } } } } }, "output_schema": { "type": "object", "required": [ "tool", "result" ], "properties": { "tool": { "type": "string" }, "result": { "type": "object", "required": [ "text", "files" ], "properties": { "text": { "type": [ "string", "null" ] }, "data": { "description": "Parsed JSON when the textual result is JSON." }, "name": { "type": "string" }, "mime": { "type": "string" }, "files": { "type": "array", "items": { "type": "object", "required": [ "name", "mime", "bytes", "base64" ], "properties": { "name": { "type": "string" }, "mime": { "type": "string" }, "bytes": { "type": "integer" }, "base64": { "type": "string", "contentEncoding": "base64" } } } } } } } }, "processing": "Server; self-hosted native engine, no external conversion service", "limits": { "request_bytes": 31457280, "input_characters": 1000000, "files_bytes": 20971520, "max_files": 1, "timeout_seconds": 90 } }, "agent_processing": "Server; self-hosted native engine, no external conversion service", "browser_agent": { "supported": true, "name": "run_current_tool", "discovery": "WebMCP on the human page in a supporting browser", "verification": "See audit/API-AUDIT.md; availability is not verification" } }, "markdown": "# Speaker diarization\n\n- Human page: https://happytails.ai/en/speech-tools/speaker-diarization/\n- Markdown: https://happytails.ai/en/speech-tools/speaker-diarization.md\n- Structured JSON: https://happytails.ai/agents/en/speech-tools/speaker-diarization/index.json\n- Visibility: public\n\n## Tool capabilities\n\n| Property | Value |\n| --- | --- |\n| Tool ID | `speaker-diarization` |\n| Category | speech-tools |\n| Implementation | Registered implementation. Registration alone is not a production-quality guarantee. |\n| Browser execution | Use the visible action or interactive controls. |\n| Browser processing | Server required |\n| HTTP API | /api/v1/tools/speaker-diarization/run |\n| Browser-agent interface | run_current_tool |\n\n### Implementation notes\n\nGroup non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.\n\n### Inputs and settings\n\nNo free-text input is required. Use the file inputs and/or settings below.\n\nAccepted browser files: `.wav,.mp3,.m4a,.flac,.mp4,.mov,.webm`. Choose one file.\n\nBrowser file-input limit: 20971520 bytes in total. HTTP limits can differ.\n\n| Setting | Meaning | Control type | Default | Constraints |\n| --- | --- | --- | --- | --- |\n| `speakers` | Number of speakers (0 to estimate) | number | 0 | min=0 max=6 step=1 |\n\n### HTTP contract\n\nAPI calls process supplied data on the server. The website origin is `https://happytails.ai`. Server API hosting is not connected on the public website yet. Use your local server origin to test these requests.\n\n```http\nPOST https://happytails.ai/api/v1/tools/speaker-diarization/run\nContent-Type: application/json\n```\n\n[Detailed operation reference](https://happytails.ai/en/api/tools/speaker-diarization.md) · [Shared API guide](https://happytails.ai/en/api.md)\n\n#### Limits\n\n| Limit | Value |\n| --- | --- |\n| request_bytes | 31457280 |\n| input_characters | 1000000 |\n| files_bytes | 20971520 |\n| max_files | 1 |\n| timeout_seconds | 90 |\n\n#### Complete input schema\n\n```json\n{\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"properties\": {\n \"input\": {\n \"type\": \"string\",\n \"maxLength\": 1000000,\n \"default\": \"\",\n \"description\": \"Plain text input. File tools use files instead unless otherwise documented.\"\n },\n \"options\": {\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"properties\": {\n \"speakers\": {\n \"type\": \"number\",\n \"description\": \"Number of speakers (0 to estimate)\",\n \"default\": 0.0,\n \"minimum\": 0,\n \"maximum\": 6,\n \"multipleOf\": 1\n }\n }\n },\n \"files\": {\n \"type\": \"array\",\n \"maxItems\": 1,\n \"items\": {\n \"type\": \"object\",\n \"required\": [\n \"name\",\n \"base64\"\n ],\n \"additionalProperties\": false,\n \"properties\": {\n \"name\": {\n \"type\": \"string\",\n \"maxLength\": 200,\n \"description\": \"Filename only, no path.\"\n },\n \"mime\": {\n \"type\": \"string\",\n \"maxLength\": 150\n },\n \"bytes\": {\n \"type\": \"integer\",\n \"minimum\": 0,\n \"description\": \"Optional decoded byte count; must match content if supplied.\"\n },\n \"base64\": {\n \"type\": \"string\",\n \"contentEncoding\": \"base64\",\n \"description\": \"File bytes as padded base64. Use the tool-specific limits.files_bytes value for the total decoded input size.\"\n }\n }\n }\n }\n }\n}\n```\n\n#### Complete output schema\n\n```json\n{\n \"type\": \"object\",\n \"required\": [\n \"tool\",\n \"result\"\n ],\n \"properties\": {\n \"tool\": {\n \"type\": \"string\"\n },\n \"result\": {\n \"type\": \"object\",\n \"required\": [\n \"text\",\n \"files\"\n ],\n \"properties\": {\n \"text\": {\n \"type\": [\n \"string\",\n \"null\"\n ]\n },\n \"data\": {\n \"description\": \"Parsed JSON when the textual result is JSON.\"\n },\n \"name\": {\n \"type\": \"string\"\n },\n \"mime\": {\n \"type\": \"string\"\n },\n \"files\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"object\",\n \"required\": [\n \"name\",\n \"mime\",\n \"bytes\",\n \"base64\"\n ],\n \"properties\": {\n \"name\": {\n \"type\": \"string\"\n },\n \"mime\": {\n \"type\": \"string\"\n },\n \"bytes\": {\n \"type\": \"integer\"\n },\n \"base64\": {\n \"type\": \"string\",\n \"contentEncoding\": \"base64\"\n }\n }\n }\n }\n }\n }\n }\n}\n```\n\n#### Error and retry handling\n\nFailures return a non-200 status and `error.code` plus `error.message`. Correct invalid input before retrying; retry server-busy responses with bounded backoff. Returned files contain Base64 data, not persistent download URLs. Decode and inspect the output before treating conversion as successful. See the shared API guide for the complete status and timeout rules.\n\n### Browser-agent workflow\n\n1. Open the human page in a browser that supports WebMCP.\n2. Discover the interface exposed by that page; use its actual schema.\n3. Select files on the page first when the tool requires files.\n4. Call `run_current_tool` with valid input and settings.\n5. Inspect the returned result or error and the displayed output. Retrieve files from the displayed download links.\n\nCalls do not automatically copy or download results. Browser permissions still apply. Interface availability alone is not proof of successful execution.\n\n### Planning specification\n\nLabel speaker turns as Speaker 1 and Speaker 2; let users merge and rename speakers; review overlap; export timestamped JSON. Does not identify people by voice.\n\nThis describes intended requirements. Use the implementation notes, interface schema and observed output to determine current support.\n\n### Complete machine-readable capability record\n\n```json\n{\n \"id\": \"speaker-diarization\",\n \"category\": \"speech-tools\",\n \"built\": true,\n \"requirements\": \"Label speaker turns as Speaker 1 and Speaker 2; let users merge and rename speakers; review overlap; export timestamped JSON. Does not identify people by voice.\",\n \"implementation\": {\n \"fields\": [\n {\n \"key\": \"speakers\",\n \"label\": \"Number of speakers (0 to estimate)\",\n \"type\": \"number\",\n \"value\": 0,\n \"min\": 0,\n \"max\": 6,\n \"step\": 1\n }\n ],\n \"files\": true,\n \"accept\": \".wav,.mp3,.m4a,.flac,.mp4,.mov,.webm\",\n \"multiple\": false,\n \"input\": false,\n \"processing\": \"Server required\",\n \"native\": true,\n \"ai\": false,\n \"maxFileBytes\": 20971520,\n \"note\": \"Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.\",\n \"extra\": true,\n \"experience\": {\n \"automatic\": false,\n \"single\": false,\n \"label\": \"Your text\",\n \"placeholder\": \"Type or paste your input…\",\n \"help\": \"Up to 1,000,000 characters. Review the settings before running.\",\n \"action\": null\n },\n \"mode\": \"explicit\",\n \"interactive\": false\n },\n \"execution\": \"POST /api/v1/tools/speaker-diarization/run\",\n \"api\": {\n \"id\": \"speaker-diarization\",\n \"name\": \"Speaker diarization\",\n \"category\": \"speech-tools\",\n \"description\": \"Group non-overlapping voices using local ECAPA speaker embeddings.\",\n \"notes\": \"Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.\",\n \"human_path\": \"/en/speech-tools/speaker-diarization/\",\n \"method\": \"POST\",\n \"endpoint\": \"/api/v1/tools/speaker-diarization/run\",\n \"schema_url\": \"/api/v1/tools/speaker-diarization\",\n \"input_schema\": {\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"properties\": {\n \"input\": {\n \"type\": \"string\",\n \"maxLength\": 1000000,\n \"default\": \"\",\n \"description\": \"Plain text input. File tools use files instead unless otherwise documented.\"\n },\n \"options\": {\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"properties\": {\n \"speakers\": {\n \"type\": \"number\",\n \"description\": \"Number of speakers (0 to estimate)\",\n \"default\": 0.0,\n \"minimum\": 0,\n \"maximum\": 6,\n \"multipleOf\": 1\n }\n }\n },\n \"files\": {\n \"type\": \"array\",\n \"maxItems\": 1,\n \"items\": {\n \"type\": \"object\",\n \"required\": [\n \"name\",\n \"base64\"\n ],\n \"additionalProperties\": false,\n \"properties\": {\n \"name\": {\n \"type\": \"string\",\n \"maxLength\": 200,\n \"description\": \"Filename only, no path.\"\n },\n \"mime\": {\n \"type\": \"string\",\n \"maxLength\": 150\n },\n \"bytes\": {\n \"type\": \"integer\",\n \"minimum\": 0,\n \"description\": \"Optional decoded byte count; must match content if supplied.\"\n },\n \"base64\": {\n \"type\": \"string\",\n \"contentEncoding\": \"base64\",\n \"description\": \"File bytes as padded base64. Use the tool-specific limits.files_bytes value for the total decoded input size.\"\n }\n }\n }\n }\n }\n },\n \"output_schema\": {\n \"type\": \"object\",\n \"required\": [\n \"tool\",\n \"result\"\n ],\n \"properties\": {\n \"tool\": {\n \"type\": \"string\"\n },\n \"result\": {\n \"type\": \"object\",\n \"required\": [\n \"text\",\n \"files\"\n ],\n \"properties\": {\n \"text\": {\n \"type\": [\n \"string\",\n \"null\"\n ]\n },\n \"data\": {\n \"description\": \"Parsed JSON when the textual result is JSON.\"\n },\n \"name\": {\n \"type\": \"string\"\n },\n \"mime\": {\n \"type\": \"string\"\n },\n \"files\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"object\",\n \"required\": [\n \"name\",\n \"mime\",\n \"bytes\",\n \"base64\"\n ],\n \"properties\": {\n \"name\": {\n \"type\": \"string\"\n },\n \"mime\": {\n \"type\": \"string\"\n },\n \"bytes\": {\n \"type\": \"integer\"\n },\n \"base64\": {\n \"type\": \"string\",\n \"contentEncoding\": \"base64\"\n }\n }\n }\n }\n }\n }\n }\n },\n \"processing\": \"Server; self-hosted native engine, no external conversion service\",\n \"limits\": {\n \"request_bytes\": 31457280,\n \"input_characters\": 1000000,\n \"files_bytes\": 20971520,\n \"max_files\": 1,\n \"timeout_seconds\": 90\n }\n },\n \"agent_processing\": \"Server; self-hosted native engine, no external conversion service\",\n \"browser_agent\": {\n \"supported\": true,\n \"name\": \"run_current_tool\",\n \"discovery\": \"WebMCP on the human page in a supporting browser\",\n \"verification\": \"See audit/API-AUDIT.md; availability is not verification\"\n }\n}\n```\n\n## Page guide\n\nGroup non-overlapping voices using local ECAPA speaker embeddings.\n\n## How to use Speaker diarization\n\n1. Choose a file in WAV, MP3, M4A, FLAC, MP4, MOV or WEBM.\n2. Review the settings, then choose Run Speaker diarization.\n3. Review the result, then use Copy result or a download link when available.\n\n## What to expect\n\nGroup non-overlapping voices using local ECAPA speaker embeddings. Group non-overlapping voices using local ECAPA speaker embeddings. Up to one minute. Returns speaker labels and approximate 250 ms timestamps. It does not identify people or separate overlapping speech. Review short turns and noisy recordings.\n" }