> ## Documentation Index
> Fetch the complete documentation index at: https://typecast.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Streaming Text To Speech
> Generate speech from text using real-time streaming, allowing audio playback to begin before the entire synthesis is complete.
This endpoint streams audio data in chunks, enabling low-latency audio playback for applications requiring immediate feedback.
**Streaming Format:**
* **WAV format**: First chunk contains WAV header (size\=0xFFFFFFFF for streaming) followed by raw PCM data. Subsequent chunks contain only PCM data.
* **MP3 format**: Each chunk contains post-processed MP3 data that can be decoded independently.
**Use Cases:**
* Conversational AI, chatbots and real-time voice assistants
* Interactive applications requiring immediate audio feedback
* Long-form content where waiting for full synthesis is impractical
**Request Parameters:**
Uses the same TTSRequest schema as the standard TTS endpoint. Set `output.audio_format` to "wav" or "mp3" to control the streaming format.
## OpenAPI
```json
{
"openapi": "3.1.0",
"info": {
"title": "Typecast API",
"x-logo": {
"url": "https://typecast.ai/_ipx/_/image/logo/tc_logo.webp"
},
"version": "0.1.2"
},
"servers": [
{
"url": "https://api.typecast.ai",
"description": "Production server"
}
],
"security": [
{
"ApiKeyAuth": []
}
],
"paths": {
"/v1/text-to-speech/stream": {
"post": {
"tags": [
"Text-to-Speech"
],
"x-mint": {
"href": "/api-reference/text-to-speech/streaming-text-to-speech"
},
"summary": "Streaming Text To Speech",
"security": [
{
"ApiKeyAuth": []
}
],
"responses": {
"200": {
"content": {
"audio/wav": {
"schema": {
"type": "string",
"format": "binary",
"description": "Chunked WAV audio stream (16-bit, mono, 32000 Hz). First chunk includes WAV header with size 0xFFFFFFFF (indicating streaming), followed by raw PCM data. Subsequent chunks contain only PCM data."
},
"example": "[Binary audio stream - WAV chunks]"
}
},
"description": "Success - Returns streaming audio data in chunks"
},
"400": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "Invalid voice_id"
}
}
},
"description": "Bad Request - Invalid parameters"
},
"401": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "Invalid API key"
}
}
},
"description": "Unauthorized - Authentication failed"
},
"402": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "Insufficient credit"
}
}
},
"description": "Payment Required - Insufficient credits"
},
"404": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "Voice not found"
}
}
},
"description": "Not Found - Voice model not available"
},
"422": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"message": "The input text contains characters or symbols that cannot be synthesized into speech. Please check your input text.",
"error_code": "TEXT_NOT_SYNTHESIZABLE"
}
}
},
"description": "Validation Error - The request is invalid or the input text cannot be synthesized.\r\nInput errors detected before streaming starts return `TEXT_NOT_SYNTHESIZABLE`. After the streaming response has started, its HTTP status can no longer be changed."
},
"429": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "Too many requests"
}
}
},
"description": "Too Many Requests - Rate limit exceeded"
},
"500": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ErrorResponse"
},
"example": {
"detail": "An unexpected error occurred"
}
}
},
"description": "Internal Server Error - Server processing failed"
}
},
"deprecated": false,
"description": "Generate speech from text using real-time streaming, allowing audio playback to begin before the entire synthesis is complete.\n\nThis endpoint streams audio data in chunks, enabling low-latency audio playback for applications requiring immediate feedback.\n\n**Streaming Format:**\n\n* **WAV format**: First chunk contains WAV header (size\\=0xFFFFFFFF for streaming) followed by raw PCM data. Subsequent chunks contain only PCM data.\n* **MP3 format**: Each chunk contains post-processed MP3 data that can be decoded independently.\n\n**Use Cases:**\n\n* Conversational AI, chatbots and real-time voice assistants\n* Interactive applications requiring immediate audio feedback\n* Long-form content where waiting for full synthesis is impractical\n\n**Request Parameters:**\nUses the same TTSRequest schema as the standard TTS endpoint. Set `output.audio_format` to \"wav\" or \"mp3\" to control the streaming format.",
"operationId": "text_to_speech_stream_v1_text_to_speech_stream_post",
"requestBody": {
"content": {
"application/json": {
"schema": {
"type": "object",
"title": "TTSRequestStreamStream",
"required": [
"voice_id",
"text",
"model"
],
"properties": {
"text": {
"type": "string",
"title": "Text",
"example": "Everything is so incredibly perfect that I feel like I'm dreaming.",
"maxLength": 2000,
"minLength": 1,
"description": "Text to convert to speech. Minimum 1 character, maximum 2000 characters. Credits consumed based on text length. Supports multiple languages including English, Korean, Japanese, and Chinese. Special characters and punctuation are handled automatically."
},
"model": {
"$ref": "#/components/schemas/TTSModel",
"example": "ssfm-v30",
"description": "Voice model to use for speech synthesis.\r\n\r\n* **ssfm-v30**: Latest model with improved prosody and additional emotion presets (recommended)\r\n* **ssfm-v21**: Stable production model with reliable quality"
},
"output": {
"type": "object",
"title": "",
"properties": {
"audio_pitch": {
"type": "integer",
"title": "Audio Pitch",
"default": 0,
"example": 0,
"maximum": 12,
"minimum": -12,
"description": "Adjusts the pitch in semitones to affect perceived gender and age: -12 (one octave lower, deeper voice), -6 (half octave lower), 0 (original pitch, default), +6 (half octave higher), +12 (one octave higher, higher voice)"
},
"audio_tempo": {
"type": "number",
"title": "Audio Tempo",
"default": 1,
"example": 1,
"maximum": 2,
"minimum": 0.5,
"description": "Controls speech speed: 0.5 (half speed, very slow and clear), 0.75 (slightly slower than normal), 1.0 (normal speaking speed, default), 1.5 (50% faster than normal), 2.0 (double speed, very fast speech)"
},
"target_lufs": {
"anyOf": [
{
"type": "number",
"maximum": 0,
"minimum": -70
},
{
"type": "null"
}
],
"title": "Target Lufs",
"example": -14,
"description": "Sets the target absolute loudness (LUFS) for streaming output audio. This normalizes generated audio to a consistent loudness regardless of the original source. Cannot be used with the `volume` parameter.\r\n\r\nRecommended values: -14 (common streaming standard), -23 (broadcast standard).\r\n"
},
"audio_format": {
"enum": [
"wav",
"mp3"
],
"type": "string",
"title": "Audio Format",
"default": "wav",
"example": "wav",
"description": "Output audio format for streaming.\r\n\r\n**WAV format:**\r\n- Uncompressed PCM audio\r\n- 16-bit depth, mono channel, **32000 Hz** sample rate\r\n- Chunked transfer: first chunk contains the WAV header (size = 0xFFFFFFFF), subsequent chunks contain raw PCM data\r\n- Recommended when you want to play audio as it arrives\r\n\r\n**MP3 format:**\r\n- Compressed MPEG Layer III audio\r\n- 320 kbps bitrate, 44100 Hz sample rate\r\n- Chunked transfer: each chunk contains independently decodable MPEG frames\r\n- Recommended for bandwidth-constrained clients\r\n"
},
"remove_silence_ms": {
"anyOf": [
{
"type": "integer",
"maximum": 1000,
"minimum": 0
},
{
"type": "null"
}
],
"title": "Remove Silence Ms",
"default": null,
"example": 100,
"description": "When enabled, shortens detected silences longer than the specified duration to that duration. The value is in milliseconds (ms). The value is the duration of silence to retain, not the amount to remove.\r\n\r\n**Accepted values:**\r\n\r\n- An integer from **0 to 1000**, with a recommended range of 0 to 200.\r\n- Omitted or `null`: silence removal is disabled.\r\n- `0`: removes detected silence longer than 0 ms. **This does not disable the feature.**\r\n- Booleans, strings, fractional values, and out-of-range values are invalid.\r\n\r\n**Example:** `100` shortens detected silences longer than 100 ms to 100 ms. Lower values include shorter silence segments for removal and leave less silence in each affected segment."
}
},
"description": "Streaming audio output settings including pitch (-12 to +12 semitones), tempo (0.5x to 2.0x), format (wav/mp3), and target\\_lufs (-70 to 0 LUFS). Note: volume is not available in streaming mode.\r\n\r\nAlso supports `remove_silence_ms` (integer, 0–1000 ms) for silence removal."
},
"prompt": {
"oneOf": [
{
"$ref": "#/components/schemas/SmartPrompt"
},
{
"$ref": "#/components/schemas/PresetPrompt"
},
{
"$ref": "#/components/schemas/Prompt"
}
],
"title": "Prompt",
"description": "Emotion and style settings for the generated speech, including emotion type (happy/sad/angry/normal) and intensity (0.0 to 2.0) to control the emotional expression"
},
"language": {
"type": "string",
"title": "Language",
"example": "eng",
"description": "Language code following ISO 639-3 standard. Case-insensitive (both \"ENG\" and \"eng\" are accepted). If not provided, will be auto-detected based on text content.\r\n\r\n\r\n ssfm-v30 Supported Languages (37)
\r\n\r\n | Code | Language | Code | Language | Code | Language |\r\n | ---- | --------- | ---- | ---------- | ---- | ---------- |\r\n | ARA | Arabic | IND | Indonesian | POR | Portuguese |\r\n | BEN | Bengali | ITA | Italian | RON | Romanian |\r\n | BUL | Bulgarian | JPN | Japanese | RUS | Russian |\r\n | CES | Czech | KOR | Korean | SLK | Slovak |\r\n | DAN | Danish | MSA | Malay | SPA | Spanish |\r\n | DEU | German | NAN | Min Nan | SWE | Swedish |\r\n | ELL | Greek | NLD | Dutch | TAM | Tamil |\r\n | ENG | English | NOR | Norwegian | TGL | Tagalog |\r\n | FIN | Finnish | PAN | Punjabi | THA | Thai |\r\n | FRA | French | POL | Polish | TUR | Turkish |\r\n | HIN | Hindi | UKR | Ukrainian | VIE | Vietnamese |\r\n | HRV | Croatian | YUE | Cantonese | ZHO | Chinese |\r\n | HUN | Hungarian | | | | |\r\n \r\n\r\n\r\n ssfm-v21 Supported Languages (27)
\r\n\r\n | Code | Language | Code | Language | Code | Language |\r\n | ---- | --------- | ---- | ---------- | ---- | --------- |\r\n | ARA | Arabic | IND | Indonesian | RON | Romanian |\r\n | BUL | Bulgarian | ITA | Italian | RUS | Russian |\r\n | CES | Czech | JPN | Japanese | SLK | Slovak |\r\n | DAN | Danish | KOR | Korean | SPA | Spanish |\r\n | DEU | German | MSA | Malay | SWE | Swedish |\r\n | ELL | Greek | NLD | Dutch | TAM | Tamil |\r\n | ENG | English | POL | Polish | TGL | Tagalog |\r\n | FIN | Finnish | POR | Portuguese | UKR | Ukrainian |\r\n | FRA | French | HRV | Croatian | ZHO | Chinese |\r\n "
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"example": "tc_60e5426de8b95f1d3000d7b5",
"description": "Voice identifier. Two prefixes are supported:\r\n\r\n* `tc_` — Built-in Typecast voices (e.g., `tc_60e5426de8b95f1d3000d7b5`). See [Listing all voices](/docs/api-reference/voices/list-voices) for available IDs.\r\n* `uc_` — Custom voices created via [Instant cloning](/docs/api-reference/voices/instant-cloning) (e.g., `uc_64a1b2c3d4e5f6a7b8c9d0e1`). Only the owner of a cloned voice can use it.\r\n\r\nCase-sensitive: must use lowercase prefix."
}
},
"description": "Text-to-speech streaming request parameters"
}
}
},
"required": true,
"description": ""
},
"x-codeSamples": [
{
"lang": "cURL",
"label": "cURL (stream + play)",
"source": "# Pipe streaming audio directly into ffplay for real-time playback.\n# Requires: ffmpeg (brew/choco/apt install ffmpeg)\ncurl -N -s --request POST \\\n --url https://api.typecast.ai/v1/text-to-speech/stream \\\n --header 'Content-Type: application/json' \\\n --header 'X-API-KEY: ' \\\n --data @- <\", \"Content-Type\": \"application/json\"}\npayload = {\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n}\n\nresp = requests.post(\n f\"{API_HOST}/v1/text-to-speech/stream\",\n headers=headers, json=payload, stream=True, timeout=60,\n)\nresp.raise_for_status()\n\nwith sd.RawOutputStream(samplerate=32000, channels=1, dtype=\"int16\") as player:\n buf, first = bytearray(), True\n for chunk in resp.iter_content(chunk_size=4096):\n if not chunk:\n continue\n if first:\n chunk = chunk[44:] # strip WAV header\n first = False\n buf.extend(chunk)\n # Write 2-byte-aligned slices (int16 samples).\n n = len(buf) - (len(buf) % 2)\n if n:\n player.write(bytes(buf[:n]))\n del buf[:n]\n\nprint(\"Playback completed\")\n"
},
{
"lang": "C#",
"label": "C# (HttpClient + ffplay)",
"source": "// Real-time playback by piping the stream into ffplay.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\nusing System;\nusing System.Diagnostics;\nusing System.Net.Http;\nusing System.Text;\nusing System.Threading.Tasks;\n\nvar client = new HttpClient();\nclient.DefaultRequestHeaders.Add(\"X-API-KEY\", \"\");\n\nvar requestBody = @\"{\n \"\"voice_id\"\": \"\"tc_60e5426de8b95f1d3000d7b5\"\",\n \"\"text\"\": \"\"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\"\",\n \"\"model\"\": \"\"ssfm-v30\"\"\n}\";\n\nvar ffplay = new Process\n{\n StartInfo = new ProcessStartInfo\n {\n FileName = \"ffplay\",\n Arguments = \"-autoexit -nodisp -loglevel error -i pipe:0\",\n RedirectStandardInput = true,\n UseShellExecute = false\n }\n};\nffplay.Start();\n\nvar request = new HttpRequestMessage(HttpMethod.Post, \"https://api.typecast.ai/v1/text-to-speech/stream\")\n{\n Content = new StringContent(requestBody, Encoding.UTF8, \"application/json\")\n};\n\n// ResponseHeadersRead enables true streaming (avoids full buffering).\nusing var response = await client.SendAsync(request, HttpCompletionOption.ResponseHeadersRead);\nresponse.EnsureSuccessStatusCode();\nusing var stream = await response.Content.ReadAsStreamAsync();\nawait stream.CopyToAsync(ffplay.StandardInput.BaseStream);\nffplay.StandardInput.Close();\nawait ffplay.WaitForExitAsync();\n"
},
{
"lang": "Kotlin",
"label": "Kotlin (OkHttp + ffplay)",
"source": "// Real-time playback by piping the OkHttp response stream into ffplay.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\n// For Android, replace the ffplay Process with AudioTrack + raw PCM feed.\nimport okhttp3.MediaType.Companion.toMediaType\nimport okhttp3.OkHttpClient\nimport okhttp3.Request\nimport okhttp3.RequestBody.Companion.toRequestBody\n\nval ffplay = ProcessBuilder(\n \"ffplay\", \"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\"\n).redirectError(ProcessBuilder.Redirect.DISCARD).start()\n\nval client = OkHttpClient()\nval body = \"\"\"\n{\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n}\n\"\"\".trimIndent().toRequestBody(\"application/json\".toMediaType())\n\nval request = Request.Builder()\n .url(\"https://api.typecast.ai/v1/text-to-speech/stream\")\n .addHeader(\"X-API-KEY\", \"\")\n .post(body)\n .build()\n\nclient.newCall(request).execute().use { response ->\n response.body?.byteStream()?.use { input -> input.copyTo(ffplay.outputStream) }\n}\nffplay.outputStream.close()\nffplay.waitFor()\n"
},
{
"lang": "C++",
"label": "C++ (libcurl + ffplay)",
"source": "// Real-time playback: libcurl write callback pipes each chunk into\n// ffplay via popen. Requires: ffmpeg (brew/choco/apt install ffmpeg)\n#include \n#include \n#include \n\nstatic FILE* player = nullptr;\n\nsize_t cb(void* ptr, size_t size, size_t nmemb, void*) {\n return fwrite(ptr, size, nmemb, player);\n}\n\nint main() {\n player = popen(\"ffplay -autoexit -nodisp -loglevel error -i pipe:0\", \"w\");\n\n CURL* curl = curl_easy_init();\n struct curl_slist* headers = nullptr;\n headers = curl_slist_append(headers, \"Content-Type: application/json\");\n headers = curl_slist_append(headers, \"X-API-KEY: \");\n\n std::string body = R\"({\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n })\";\n\n curl_easy_setopt(curl, CURLOPT_URL, \"https://api.typecast.ai/v1/text-to-speech/stream\");\n curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);\n curl_easy_setopt(curl, CURLOPT_POSTFIELDS, body.c_str());\n curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, cb);\n\n curl_easy_perform(curl);\n\n curl_slist_free_all(headers);\n curl_easy_cleanup(curl);\n pclose(player);\n return 0;\n}\n"
},
{
"lang": "C",
"label": "C (libcurl + ffplay)",
"source": "/* Real-time playback: libcurl write callback pipes each chunk into\n * ffplay via popen. Requires: ffmpeg (brew/choco/apt install ffmpeg) */\n#include \n#include \n\nstatic FILE* player = NULL;\n\nsize_t cb(void* ptr, size_t size, size_t nmemb, void* ud) {\n (void)ud;\n return fwrite(ptr, size, nmemb, player);\n}\n\nint main(void) {\n player = popen(\"ffplay -autoexit -nodisp -loglevel error -i pipe:0\", \"w\");\n\n curl_global_init(CURL_GLOBAL_ALL);\n CURL* curl = curl_easy_init();\n\n struct curl_slist* headers = NULL;\n headers = curl_slist_append(headers, \"Content-Type: application/json\");\n headers = curl_slist_append(headers, \"X-API-KEY: \");\n\n const char* body =\n \"{\"\n \"\\\"voice_id\\\":\\\"tc_60e5426de8b95f1d3000d7b5\\\",\"\n \"\\\"text\\\":\\\"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\\\",\"\n \"\\\"model\\\":\\\"ssfm-v30\\\"\"\n \"}\";\n\n curl_easy_setopt(curl, CURLOPT_URL, \"https://api.typecast.ai/v1/text-to-speech/stream\");\n curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);\n curl_easy_setopt(curl, CURLOPT_POSTFIELDS, body);\n curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, cb);\n\n curl_easy_perform(curl);\n\n curl_slist_free_all(headers);\n curl_easy_cleanup(curl);\n curl_global_cleanup();\n pclose(player);\n return 0;\n}\n"
},
{
"lang": "Swift",
"label": "Swift (URLSession + ffplay)",
"source": "// Real-time playback (macOS): pipe URLSession bytes into ffplay via\n// Process. Requires: ffmpeg (brew install ffmpeg).\n// Requires iOS 15 / macOS 12 for URLSession.bytes(for:).\n// Compile with: swiftc -parse-as-library main.swift -o streaming_tts\n// For iOS, replace Process/ffplay with AVAudioEngine + scheduled PCM buffers.\nimport Foundation\n\n@main\nstruct StreamingTTS {\n static func main() async throws {\n let ffplay = Process()\n ffplay.executableURL = URL(fileURLWithPath: \"/usr/bin/env\")\n ffplay.arguments = [\"ffplay\", \"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\"]\n let pipe = Pipe()\n ffplay.standardInput = pipe\n try ffplay.run()\n\n var request = URLRequest(url: URL(string: \"https://api.typecast.ai/v1/text-to-speech/stream\")!)\n request.httpMethod = \"POST\"\n request.setValue(\"application/json\", forHTTPHeaderField: \"Content-Type\")\n request.setValue(\"\", forHTTPHeaderField: \"X-API-KEY\")\n request.httpBody = try JSONSerialization.data(withJSONObject: [\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n ])\n\n let (bytes, _) = try await URLSession.shared.bytes(for: request)\n var buffer = Data()\n buffer.reserveCapacity(4096)\n for try await byte in bytes {\n buffer.append(byte)\n if buffer.count >= 4096 {\n try pipe.fileHandleForWriting.write(contentsOf: buffer)\n buffer.removeAll(keepingCapacity: true)\n }\n }\n if !buffer.isEmpty {\n try pipe.fileHandleForWriting.write(contentsOf: buffer)\n }\n try pipe.fileHandleForWriting.close()\n ffplay.waitUntilExit()\n }\n}\n"
},
{
"lang": "Rust",
"label": "Rust (reqwest + ffplay)",
"source": "// Real-time playback: pipe reqwest stream into ffplay via tokio Command.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\n// Cargo.toml:\n// reqwest = { version = \"0.12\", features = [\"json\", \"stream\"] }\n// tokio = { version = \"1\", features = [\"full\"] }\n// serde_json = \"1\"\nuse reqwest;\nuse serde_json::json;\nuse std::process::Stdio;\nuse tokio::io::AsyncWriteExt;\nuse tokio::process::Command;\n\n#[tokio::main]\nasync fn main() -> Result<(), Box> {\n let mut ffplay = Command::new(\"ffplay\")\n .args([\"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\"])\n .stdin(Stdio::piped())\n .spawn()?;\n let mut stdin = ffplay.stdin.take().expect(\"failed to open ffplay stdin\");\n\n let client = reqwest::Client::new();\n let body = json!({\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n });\n\n let mut response = client\n .post(\"https://api.typecast.ai/v1/text-to-speech/stream\")\n .header(\"X-API-KEY\", \"\")\n .header(\"Content-Type\", \"application/json\")\n .json(&body)\n .send()\n .await?;\n\n while let Some(chunk) = response.chunk().await? {\n stdin.write_all(&chunk).await?;\n }\n drop(stdin);\n ffplay.wait().await?;\n Ok(())\n}\n"
},
{
"lang": "JavaScript",
"label": "JavaScript (Node.js + ffplay)",
"source": "// Node 18+ (built-in fetch). Pipe streamed audio into ffplay.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\nimport { spawn } from \"node:child_process\";\n\nconst ffplay = spawn(\n \"ffplay\",\n [\"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\"],\n { stdio: [\"pipe\", \"ignore\", \"ignore\"] },\n);\n\nconst response = await fetch(\"https://api.typecast.ai/v1/text-to-speech/stream\", {\n method: \"POST\",\n headers: {\n \"Content-Type\": \"application/json\",\n \"X-API-KEY\": \"\"\n },\n body: JSON.stringify({\n voice_id: \"tc_60e5426de8b95f1d3000d7b5\",\n text: \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n model: \"ssfm-v30\"\n })\n});\nif (!response.ok) throw new Error(`HTTP ${response.status}`);\n\n// fetch().body is a Web ReadableStream — read chunks as they arrive.\nconst reader = response.body.getReader();\nwhile (true) {\n const { value, done } = await reader.read();\n if (done) break;\n ffplay.stdin.write(value);\n}\nffplay.stdin.end();\nawait new Promise((resolve) => ffplay.on(\"close\", resolve));\n"
},
{
"lang": "PHP",
"label": "PHP (curl + ffplay)",
"source": " \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\" => \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\" => \"ssfm-v30\"\n]);\n\n$ch = curl_init(\"https://api.typecast.ai/v1/text-to-speech/stream\");\ncurl_setopt_array($ch, [\n CURLOPT_POST => true,\n CURLOPT_HTTPHEADER => [\n \"Content-Type: application/json\",\n \"X-API-KEY: \"\n ],\n CURLOPT_POSTFIELDS => $payload,\n CURLOPT_WRITEFUNCTION => function ($ch, $data) use ($ffplay) {\n fwrite($ffplay, $data);\n return strlen($data);\n }\n]);\ncurl_exec($ch);\npclose($ffplay);\n"
},
{
"lang": "Go",
"label": "Go (net/http + ffplay)",
"source": "// Pipes the streaming response body into ffplay's stdin.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\npackage main\n\nimport (\n \"bytes\"\n \"io\"\n \"net/http\"\n \"os/exec\"\n)\n\nfunc main() {\n ffplay := exec.Command(\"ffplay\", \"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\")\n stdin, _ := ffplay.StdinPipe()\n if err := ffplay.Start(); err != nil {\n panic(err)\n }\n\n body := []byte(`{\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n }`)\n\n req, _ := http.NewRequest(\"POST\", \"https://api.typecast.ai/v1/text-to-speech/stream\", bytes.NewReader(body))\n req.Header.Set(\"Content-Type\", \"application/json\")\n req.Header.Set(\"X-API-KEY\", \"\")\n\n resp, err := http.DefaultClient.Do(req)\n if err != nil {\n panic(err)\n }\n defer resp.Body.Close()\n\n io.Copy(stdin, resp.Body)\n stdin.Close()\n ffplay.Wait()\n}\n"
},
{
"lang": "Java",
"label": "Java (HttpClient + ffplay)",
"source": "// Java 11+ HttpClient with InputStream body handler.\n// Pipes the streaming response into ffplay's stdin.\n// Requires: ffmpeg (brew/choco/apt install ffmpeg)\nimport java.net.URI;\nimport java.net.http.HttpClient;\nimport java.net.http.HttpRequest;\nimport java.net.http.HttpResponse;\nimport java.io.InputStream;\nimport java.io.OutputStream;\n\npublic class StreamingTTS {\n public static void main(String[] args) throws Exception {\n Process ffplay = new ProcessBuilder(\n \"ffplay\", \"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\")\n .redirectError(ProcessBuilder.Redirect.DISCARD)\n .start();\n\n String body = \"\"\"\n {\n \"voice_id\": \"tc_60e5426de8b95f1d3000d7b5\",\n \"text\": \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n \"model\": \"ssfm-v30\"\n }\n \"\"\";\n\n HttpRequest request = HttpRequest.newBuilder()\n .uri(URI.create(\"https://api.typecast.ai/v1/text-to-speech/stream\"))\n .header(\"Content-Type\", \"application/json\")\n .header(\"X-API-KEY\", \"\")\n .POST(HttpRequest.BodyPublishers.ofString(body))\n .build();\n\n HttpResponse response = HttpClient.newHttpClient()\n .send(request, HttpResponse.BodyHandlers.ofInputStream());\n\n try (InputStream in = response.body();\n OutputStream out = ffplay.getOutputStream()) {\n in.transferTo(out);\n }\n ffplay.waitFor();\n }\n}\n"
},
{
"lang": "Ruby",
"label": "Ruby (net/http + ffplay)",
"source": "# Pipes the streaming response into ffplay via IO.popen.\n# Requires: ffmpeg (brew/choco/apt install ffmpeg)\nrequire \"net/http\"\nrequire \"uri\"\nrequire \"json\"\n\nffplay = IO.popen(\n [\"ffplay\", \"-autoexit\", \"-nodisp\", \"-loglevel\", \"error\", \"-i\", \"pipe:0\"],\n \"wb\",\n)\n\nuri = URI(\"https://api.typecast.ai/v1/text-to-speech/stream\")\nhttp = Net::HTTP.new(uri.host, uri.port)\nhttp.use_ssl = true\n\nreq = Net::HTTP::Post.new(uri)\nreq[\"Content-Type\"] = \"application/json\"\nreq[\"X-API-KEY\"] = \"\"\nreq.body = {\n voice_id: \"tc_60e5426de8b95f1d3000d7b5\",\n text: \"Thanks for reaching out. Your reservation has been confirmed for Friday at 7 PM.\",\n model: \"ssfm-v30\"\n}.to_json\n\nhttp.request(req) do |response|\n response.read_body { |chunk| ffplay.write(chunk) }\nend\n\nffplay.close\n"
},
{
"lang": "cURL",
"label": "Silence removal",
"source": "curl --no-buffer --request POST 'https://api.typecast.ai/v1/text-to-speech/stream' \\\n --header 'X-API-KEY: ' \\\n --header 'Content-Type: application/json' \\\n --output review.wav \\\n --data-binary @- <<'JSON'\n{\n \"voice_id\": \"\",\n \"text\": \"Hello. Thank you for listening.\",\n \"model\": \"ssfm-v30\",\n \"output\": {\n \"audio_format\": \"wav\",\n \"remove_silence_ms\": 300\n }\n}\nJSON\n"
}
]
}
}
},
"components": {
"schemas": {
"Limits": {
"type": "object",
"title": "Limits",
"required": [
"concurrency_limit"
],
"properties": {
"concurrency_limit": {
"type": "integer",
"title": "Concurrency Limit",
"description": "Maximum number of concurrent requests allowed"
},
"custom_voice_slot": {
"type": "integer",
"title": "Custom Voice Slot",
"default": 0,
"minimum": 0,
"description": "Maximum number of custom voices (created via instant cloning) allowed on the current plan."
},
"professional_voice_slot": {
"type": "integer",
"title": "Professional Voice Slot",
"default": 0,
"minimum": 0,
"description": "프리미엄 클로닝 보이스 슬롯 한도"
}
},
"description": "Usage limit information"
},
"Prompt": {
"title": "Prompt (ssfm-v21)",
"properties": {
"emotion_preset": {
"example": "normal",
"description": "Emotion preset to apply.\r\n\r\nSupported emotions for ssfm-v21: normal, happy, sad, angry\r\n\r\nCheck available emotions for each voice in the `models` field of the `GET /v3/voices` API response.\r\n"
},
"emotion_intensity": {
"example": 1,
"description": "Controls the strength of emotional expression (0.0 to 2.0).\r\n\r\n- 0.0: Completely neutral\r\n- 1.0: Standard expression (default)\r\n- 2.0: Maximum intensity\r\n"
}
},
"description": "Emotion and style settings for the generated speech."
},
"AgeEnum": {
"enum": [
"child",
"teenager",
"young_adult",
"middle_age",
"elder"
],
"type": "string",
"title": "AgeEnum",
"description": "Age group classification enum - Converts database values (Korean) to API values (English).\n\nAvailable values:\n- **child**: Child voice (under 12 years old)\n- **teenager**: Teenage voice (13-19 years old)\n- **young_adult**: Young adult voice (20-35 years old)\n- **middle_age**: Middle-aged voice (36-60 years old)\n- **elder**: Elder voice (over 60 years old)\n"
},
"Credits": {
"type": "object",
"title": "Credits",
"required": [
"plan_credits",
"used_credits"
],
"properties": {
"plan_credits": {
"type": "integer",
"title": "Plan Credits",
"description": "Total monthly credits provided by the plan"
},
"used_credits": {
"type": "integer",
"title": "Used Credits",
"description": "Number of credits used"
}
},
"description": "Credit usage information"
},
"VoiceV2": {
"type": "object",
"title": "VoiceV2",
"required": [
"voice_id",
"voice_name",
"models",
"voice_type"
],
"properties": {
"age": {
"anyOf": [
{
"$ref": "#/components/schemas/AgeEnum"
},
{
"type": "null"
}
],
"description": "Voice age group classification (child/teenager/young_adult/middle_age/elder)"
},
"gender": {
"anyOf": [
{
"$ref": "#/components/schemas/GenderEnum"
},
{
"type": "null"
}
],
"description": "Voice gender classification (male/female)"
},
"models": {
"type": "array",
"items": {
"$ref": "#/components/schemas/ModelInfo"
},
"title": "Models",
"description": "List of supported TTS models with their available emotions (e.g., [{'version': 'ssfm-v21', 'emotions': ['happy', 'sad']}])"
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"description": "Unique voice identifier. Built-in voices use the `tc_` prefix (e.g., `tc_60e5426de8b95f1d3000d7b5`); cloned custom voices created via `POST /v1/voices/clone` use the `uc_` prefix. Use `/v3/voices` to list the built-in and completed custom voices available to your account."
},
"use_cases": {
"type": "array",
"items": {
"type": "string"
},
"title": "Use Cases",
"description": "List of use case categories this voice is suitable for"
},
"voice_name": {
"type": "string",
"title": "Voice Name",
"description": "Human-readable name of the voice"
},
"voice_type": {
"$ref": "#/components/schemas/VoiceType",
"description": "Voice type — `original` for Typecast-provided stock voices, `custom` for user-cloned voices."
}
},
"description": "V2 Voice response model with model-grouped emotions and enhanced metadata"
},
"VoiceV3": {
"type": "object",
"title": "VoiceV3",
"required": [
"voice_id",
"voice_name",
"voice_type",
"models"
],
"properties": {
"age": {
"anyOf": [
{
"$ref": "#/components/schemas/AgeEnum"
},
{
"type": "null"
}
],
"description": "Voice age group."
},
"gender": {
"anyOf": [
{
"$ref": "#/components/schemas/GenderEnum"
},
{
"type": "null"
}
],
"description": "Voice gender (`male` or `female`)."
},
"models": {
"type": "array",
"items": {
"$ref": "#/components/schemas/ModelInfo"
},
"title": "Models",
"description": "Supported TTS models and their available emotions."
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"description": "Unique voice identifier. Built-in voices use the `tc_` prefix and custom voices use the `uc_` prefix."
},
"use_cases": {
"type": "array",
"items": {
"type": "string"
},
"title": "Use Cases",
"description": "Recommended use-case categories."
},
"voice_name": {
"$ref": "#/components/schemas/VoiceName",
"description": "Localized voice names, such as `{\"eng\": \"Daejin\", \"kor\": \"대진\"}`."
},
"voice_type": {
"$ref": "#/components/schemas/VoiceType",
"description": "Voice type (`original` or `custom`)."
},
"preview_url": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Preview Url",
"description": "Preview audio URL when available."
}
},
"description": "Voice metadata with localized names and a preview audio URL."
},
"PlanTier": {
"enum": [
"free",
"lite",
"plus",
"custom"
],
"type": "string",
"title": "PlanTier",
"description": "Subscription plan for the API.\n\nAvailable values:\n- **free**: Free tier with limited credits\n- **lite**: Lite plan with moderate credits\n- **plus**: Plus plan with higher credits\n- **custom**: Custom enterprise plan"
},
"TTSModel": {
"enum": [
"ssfm-v30",
"ssfm-v21"
],
"type": "string",
"title": "TTSModel",
"description": "TTS model version to use for speech synthesis. Different models offer varying capabilities and quality levels.\n\nAvailable models:\n- **ssfm-v30**: Latest model with improved prosody and additional emotion presets (recommended)\n- **ssfm-v21**: Stable production model with proven reliability and consistent quality\n"
},
"ModelInfo": {
"type": "object",
"title": "ModelInfo",
"required": [
"version",
"emotions"
],
"properties": {
"version": {
"$ref": "#/components/schemas/TTSModel",
"description": "TTS model version (e.g., ssfm-v21, ssfm-v30)"
},
"emotions": {
"type": "array",
"items": {
"type": "string"
},
"title": "Emotions",
"description": "List of supported emotions for this model"
}
},
"description": "Model information including version and supported emotions"
},
"VoiceName": {
"type": "object",
"title": "VoiceName",
"required": [
"eng",
"kor"
],
"properties": {
"eng": {
"type": "string",
"title": "Eng",
"description": "English voice name."
},
"kor": {
"type": "string",
"title": "Kor",
"description": "Korean voice name."
}
},
"description": "Localized voice names keyed by ISO 639-3 language code."
},
"VoiceType": {
"enum": [
"original",
"custom"
],
"type": "string",
"title": "VoiceType",
"description": "Voice type classification.\n\n- `original` — Typecast-provided stock voices available to every account.\n- `custom` — Voices the user created by uploading or cloning their own sample."
},
"GenderEnum": {
"enum": [
"male",
"female"
],
"type": "string",
"title": "GenderEnum",
"description": "Gender classification enum - Converts database values (Korean) to API values (English).\n\nAvailable values:\n- **male**: Male voice\n- **female**: Female voice\n"
},
"EmotionEnum": {
"enum": [
"normal",
"sad",
"happy",
"angry",
"whisper",
"toneup",
"tonedown"
],
"type": "string",
"title": "EmotionEnum",
"description": "Available emotion presets for speech synthesis. Each emotion affects the tone, pace, and expressiveness of the generated speech.\n\n**ssfm-v21 Supported Emotions (4 types):**\n- normal: Neutral, balanced tone\n- happy: Bright, cheerful expression\n- sad: Melancholic, subdued tone\n- angry: Strong, intense delivery\n\n**ssfm-v30 Supported Emotions (7 types):**\n- normal: Neutral, balanced tone\n- happy: Bright, cheerful expression\n- sad: Melancholic, subdued tone\n- angry: Strong, intense delivery\n- whisper: Soft, quiet speech\n- toneup: Higher tonal emphasis\n- tonedown: Lower tonal emphasis\n\nCheck available emotions for each voice in the `models` field of the `GET /v3/voices` API response.\n"
},
"SmartPrompt": {
"type": "object",
"title": "SmartPrompt (ssfm-v30)",
"example": {
"next_text": "I am literally bursting with happiness and I never want this feeling to end!",
"emotion_type": "smart",
"previous_text": "I feel like I'm walking on air and I just want to scream with joy!"
},
"properties": {
"next_text": {
"type": "string",
"title": "Next Text",
"default": "",
"example": "I am literally bursting with happiness and I never want this feeling to end!",
"description": "Text that comes AFTER the `text` field in TTSRequest. Provides forward context for emotion inference.\r\n\r\nThe model analyzes the flow: `previous_text` → `text` (synthesized) → `next_text`\r\n\r\n- Maximum 2000 characters\r\n- Helps the model anticipate emotional transitions\r\n- Leave empty if no following context is available\r\n"
},
"emotion_type": {
"type": "string",
"const": "smart",
"title": "Emotion Type",
"default": "smart",
"description": "Discriminator field to identify the prompt type. Must be set to \"smart\" for context-aware emotion inference.\r\n"
},
"previous_text": {
"type": "string",
"title": "Previous Text",
"default": "",
"example": "I feel like I'm walking on air and I just want to scream with joy!",
"description": "Text that comes BEFORE the `text` field in TTSRequest. Provides backward context for emotion inference.\r\n\r\nThe model analyzes the flow: `previous_text` → `text` (synthesized) → `next_text`\r\n\r\n- Maximum 2000 characters\r\n- Helps the model understand emotional build-up and context\r\n- Leave empty if no preceding context is available\r\n"
}
},
"description": "Emotion and style settings for the generated speech.",
"additionalProperties": false
},
"PresetPrompt": {
"type": "object",
"title": "PresetPrompt (ssfm-v30)",
"properties": {
"emotion_type": {
"type": "string",
"const": "preset",
"title": "Emotion Type",
"default": "preset",
"description": "Discriminator field to identify the prompt type. Must be set to \"preset\" for preset-based emotion control.\r\n"
},
"emotion_preset": {
"$ref": "#/components/schemas/EmotionEnum",
"default": "normal",
"example": "normal",
"description": "Emotion preset to apply to the generated speech.\r\n\r\nSupported emotions: normal, happy, sad, angry, whisper, toneup, tonedown\r\n\r\nCheck available emotions for each voice in the `models` field of the `GET /v3/voices` API response.\r\n"
},
"emotion_intensity": {
"type": "number",
"title": "Emotion Intensity",
"default": 1,
"example": 1,
"maximum": 2,
"minimum": 0,
"description": "Controls the strength of emotional expression in the generated speech.\r\n\r\n- 0.0: Completely neutral, no emotional coloring\r\n- 0.5: Subtle emotional hints\r\n- 1.0: Standard emotional expression (default)\r\n- 1.5: Strong emotional emphasis\r\n- 2.0: Maximum intensity, highly expressive\r\n"
}
},
"description": "Emotion and style settings for the generated speech.",
"additionalProperties": false
},
"UseCasesEnum": {
"enum": [
"Announcer",
"Anime",
"Audiobook",
"Conversational",
"Documentary",
"E-learning",
"Rapper",
"Game",
"Tiktok/Reels",
"News",
"Podcast",
"Voicemail",
"Ads"
],
"type": "string",
"description": "Voice use case categories for content type filtering. Each voice is tagged with one or more use cases indicating its suitability for specific content types.\n\n**Available Use Cases:**\n- **Announcer**: Public announcements and presentations\n- **Anime**: Animation and character voices\n- **Audiobook**: Long-form narration and storytelling\n- **Conversational**: Chatbots and conversational AI\n- **Documentary**: Documentary narration and commentary\n- **E-learning**: Educational content and tutorials\n- **Rapper**: Rap and music performance\n- **Game**: Video game characters and narration\n- **Tiktok/Reels**: Short-form social media content\n- **News**: News broadcasting\n- **Podcast**: Broadcasting and podcast production\n- **Voicemail**: IVR systems and voice assistants\n- **Ads**: Advertising and promotional content\n"
},
"ErrorResponse": {
"type": "object",
"example": {
"message": "The input text contains characters or symbols that cannot be synthesized into speech. Please check your input text.",
"error_code": "TEXT_NOT_SYNTHESIZABLE"
},
"properties": {
"detail": {
"type": "string",
"description": "Error message describing the issue"
},
"message": {
"type": "string",
"description": "Human-readable message for structured errors"
},
"error_code": {
"type": "string",
"description": "Machine-readable error code for structured errors"
}
},
"description": "API errors use `detail` for legacy and validation errors, or `error_code` and `message` for structured errors."
},
"CustomVoiceItem": {
"type": "object",
"title": "CustomVoiceItem",
"required": [
"voice_id",
"name",
"model",
"source",
"status",
"created_at"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"description": "Voice name."
},
"error": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"title": "Error",
"description": "Failure reason when `status` is `failed`."
},
"model": {
"type": "string",
"title": "Model",
"description": "TTS model version."
},
"source": {
"type": "string",
"title": "Source",
"description": "Creation method (`instant` or `professional`)."
},
"status": {
"$ref": "#/components/schemas/CustomVoiceStatus",
"description": "Current creation or training status."
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"description": "Unique custom voice identifier with the `uc_` prefix."
},
"created_at": {
"type": "string",
"title": "Created At",
"format": "date-time",
"description": "Creation time in UTC."
}
},
"description": "A custom voice owned by the authenticated account."
},
"ValidationError": {
"type": "object",
"title": "ValidationError",
"required": [
"loc",
"msg",
"type"
],
"properties": {
"ctx": {
"type": "object",
"title": "Context"
},
"loc": {
"type": "array",
"items": {
"anyOf": [
{
"type": "string"
},
{
"type": "integer"
}
]
},
"title": "Location"
},
"msg": {
"type": "string",
"title": "Message"
},
"type": {
"type": "string",
"title": "Error Type"
},
"input": {
"title": "Input"
}
}
},
"RecommendedVoice": {
"type": "object",
"title": "RecommendedVoice",
"required": [
"voice_id",
"voice_name",
"score"
],
"properties": {
"score": {
"type": "number",
"title": "Score",
"example": 0.92,
"description": "Recommendation relevance score. Higher values indicate a stronger match for the query."
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"example": "tc_60e5426de8b95f1d3000d7b5",
"description": "Typecast voice identifier with the `tc_` prefix. Use this value as `voice_id` in text-to-speech requests."
},
"voice_name": {
"type": "string",
"title": "Voice Name",
"example": "Olivia",
"description": "Human-readable voice name."
}
},
"description": "Recommended voice candidate returned by `GET /v1/voices/recommendations`, sorted by relevance."
},
"CustomVoiceStatus": {
"enum": [
"pending",
"training",
"completed",
"failed"
],
"type": "string",
"title": "CustomVoiceStatus",
"description": "Current creation or training status of a custom voice."
},
"CustomVoiceResponse": {
"type": "object",
"title": "CustomVoiceResponse",
"required": [
"voice_id",
"name",
"model",
"status"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"description": "Human-readable voice name (1-30 characters)."
},
"model": {
"type": "string",
"allOf": [
{
"$ref": "#/components/schemas/TTSModel"
}
],
"title": "Model",
"description": "Engine model the voice was cloned for (`ssfm-v21` or `ssfm-v30`)."
},
"status": {
"$ref": "#/components/schemas/CustomVoiceStatus",
"description": "생성/학습 상태"
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"example": "uc_64a1b2c3d4e5f6a7b8c9d0e1",
"description": "Custom voice identifier with the `uc_` prefix. Use this value as `voice_id` in `POST /v1/text-to-speech` and other endpoints that accept `voice_id`."
}
},
"description": "Response of `POST /v1/voices/clone` — custom voice metadata returned by instant cloning."
},
"HTTPValidationError": {
"type": "object",
"title": "HTTPValidationError",
"properties": {
"detail": {
"type": "array",
"items": {
"$ref": "#/components/schemas/ValidationError"
},
"title": "Detail"
}
}
},
"PauseComposeSegment": {
"type": "object",
"title": "PauseComposeSegment",
"examples": [
{
"type": "pause",
"duration_seconds": 1.5
}
],
"required": [
"type",
"duration_seconds"
],
"properties": {
"type": {
"type": "string",
"const": "pause",
"title": "Type",
"default": "pause",
"description": "Segment discriminator. Always `pause`."
},
"duration_seconds": {
"type": "number",
"title": "Duration Seconds",
"example": 1.5,
"maximum": 10,
"description": "Pause duration in seconds. Each pause may be up to 10 seconds; all pauses combined may be up to 60 seconds.",
"exclusiveMinimum": 0
}
},
"description": "A silent interval inserted without consuming credits.",
"additionalProperties": false
},
"AlignmentSegmentWord": {
"type": "object",
"title": "AlignmentSegmentWord",
"required": [
"text",
"start",
"end"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"description": "End time of this segment, in seconds from the beginning of the audio."
},
"text": {
"type": "string",
"title": "Text",
"description": "The text fragment from the original transcript (includes any attached punctuation)."
},
"start": {
"type": "number",
"title": "Start",
"description": "Start time of this segment, in seconds from the beginning of the audio."
}
},
"description": "A single word-level alignment segment between the original transcript and the generated audio."
},
"SubscriptionResponse": {
"type": "object",
"title": "SubscriptionResponse",
"required": [
"plan",
"credits",
"limits",
"professional_clone_generations"
],
"properties": {
"plan": {
"$ref": "#/components/schemas/PlanTier",
"description": "Current subscription plan"
},
"limits": {
"$ref": "#/components/schemas/Limits",
"description": "Usage limit information"
},
"credits": {
"$ref": "#/components/schemas/Credits",
"description": "Credit usage information"
},
"professional_clone_generations": {
"$ref": "#/components/schemas/ProfessionalCloneGenerations"
}
},
"description": "Subscription information response model"
},
"AlignmentSegmentCharacter": {
"type": "object",
"title": "AlignmentSegmentCharacter",
"required": [
"text",
"start",
"end"
],
"properties": {
"end": {
"type": "number",
"title": "End",
"description": "End time of this segment, in seconds from the beginning of the audio."
},
"text": {
"type": "string",
"title": "Text",
"description": "The text fragment from the original transcript (includes any attached punctuation and whitespace)."
},
"start": {
"type": "number",
"title": "Start",
"description": "Start time of this segment, in seconds from the beginning of the audio."
}
},
"description": "A single character-level alignment segment between the original transcript and the generated audio."
},
"CustomVoiceCreateResponse": {
"type": "object",
"title": "CustomVoiceCreateResponse",
"required": [
"voice_id",
"name",
"model",
"status"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"description": "Voice name."
},
"model": {
"type": "string",
"title": "Model",
"description": "TTS model version."
},
"status": {
"$ref": "#/components/schemas/CustomVoiceStatus",
"description": "Current creation or training status."
},
"voice_id": {
"type": "string",
"title": "Voice Id",
"description": "Unique custom voice identifier with the `uc_` prefix."
}
},
"description": "Result returned after creating an instant or professional custom voice."
},
"TTSWithTimestampsResponse": {
"type": "object",
"title": "TTSWithTimestampsResponse",
"required": [
"audio",
"audio_format",
"audio_duration",
"words",
"characters"
],
"properties": {
"audio": {
"type": "string",
"title": "Audio",
"description": "Base64-encoded audio bytes. Decode and write to a file using the `audio_format` extension."
},
"words": {
"anyOf": [
{
"type": "array",
"items": {
"$ref": "#/components/schemas/AlignmentSegmentWord"
}
},
{
"type": "null"
}
],
"title": "Words",
"description": "Word-level timestamps (with attached punctuation). `null` when the request uses `granularity=char`."
},
"characters": {
"anyOf": [
{
"type": "array",
"items": {
"$ref": "#/components/schemas/AlignmentSegmentCharacter"
}
},
{
"type": "null"
}
],
"title": "Characters",
"description": "Character-level timestamps (including punctuation and whitespace). `null` when the request uses `granularity=word`."
},
"audio_format": {
"enum": [
"wav",
"mp3"
],
"type": "string",
"title": "Audio Format",
"description": "Audio encoding format of the bytes in `audio` — either `wav` or `mp3`, mirroring the request's `output.audio_format`."
},
"audio_duration": {
"type": "number",
"title": "Audio Duration",
"description": "Length of the generated audio in seconds."
}
},
"description": "Response payload for POST /v1/text-to-speech/with-timestamps — base64-encoded audio plus per-word and per-character timestamps aligned with the generated speech."
},
"ProfessionalCloneGenerations": {
"type": "object",
"title": "ProfessionalCloneGenerations",
"properties": {
"plan_generations": {
"type": "integer",
"title": "Plan Generations",
"default": 0,
"minimum": 0,
"description": "플랜 제공 학습 횟수"
},
"used_generations": {
"type": "integer",
"title": "Used Generations",
"default": 0,
"minimum": 0,
"description": "사용된 학습 횟수"
}
},
"description": "프리미엄 클로닝 학습 횟수 정보"
},
"Body_create_voice_clone_v1_voices_clone_post": {
"type": "object",
"title": "Body_create_voice_clone_v1_voices_clone_post",
"required": [
"file",
"name",
"model"
],
"properties": {
"file": {
"type": "string",
"title": "File",
"format": "binary",
"description": "Audio sample. WAV or MP3, max 25 MB, 5-150 seconds.",
"contentMediaType": "application/octet-stream"
},
"name": {
"type": "string",
"title": "Name",
"maxLength": 30,
"minLength": 1,
"description": "Voice name (1-30 characters)."
},
"model": {
"$ref": "#/components/schemas/TTSModel",
"allOf": [
{
"$ref": "#/components/schemas/TTSModel"
}
],
"description": "Engine model the voice is cloned for (`ssfm-v21` or `ssfm-v30`)."
}
},
"description": "Multipart request body for instant cloning."
},
"Body_create_instant_clone_v1_custom_voices_instant_clone_post": {
"type": "object",
"title": "Body_create_instant_clone_v1_custom_voices_instant_clone_post",
"required": [
"file",
"name",
"model"
],
"properties": {
"file": {
"type": "string",
"title": "File",
"format": "binary",
"description": "One WAV or MP3 recording, 25 MiB or less and 5 to 150 seconds long.",
"contentMediaType": "application/octet-stream"
},
"name": {
"type": "string",
"title": "Name",
"example": "Product Narrator",
"maxLength": 30,
"minLength": 1,
"description": "Voice name, up to 30 characters."
},
"model": {
"$ref": "#/components/schemas/TTSModel",
"example": "ssfm-v30",
"description": "TTS model version."
}
},
"description": "Multipart request body for instant voice cloning."
},
"Body_create_professional_clone_v1_custom_voices_professional_clone_post": {
"type": "object",
"title": "Body_create_professional_clone_v1_custom_voices_professional_clone_post",
"required": [
"name",
"files",
"model",
"language"
],
"properties": {
"name": {
"type": "string",
"title": "Name",
"example": "Custom Voice Name",
"maxLength": 30,
"minLength": 1,
"description": "Voice name, up to 30 characters."
},
"files": {
"type": "array",
"items": {
"type": "string",
"format": "binary",
"contentMediaType": "application/octet-stream"
},
"title": "Files",
"description": "WAV or MP3 recordings used for training. Only one file can be uploaded.\r\n\r\n**Upload limits**\r\n\r\n* One WAV or MP3 file\r\n* File size: 1 GiB or less\r\n* Duration: 5 minutes to 3 hours\r\n* Sample rate: 16 kHz or higher\r\n\r\nFor best results, we recommend audio that meets the following conditions:\r\n\r\n* Record in a speaking style that closely matches how you want the generated voice to sound.\r\n* Record in a quiet environment without background noise.\r\n* Include only one speaker.\r\n* Record in the language specified in the `language` field.\r\n* Longer input audio results in higher-quality generated voices."
},
"model": {
"$ref": "#/components/schemas/TTSModel",
"example": "ssfm-v30",
"description": "TTS model version."
},
"language": {
"type": "string",
"title": "Language",
"example": "eng",
"description": "ISO 639-3 language code, such as `kor` or `eng`."
}
},
"description": "Multipart request body for professional voice cloning."
}
},
"securitySchemes": {
"ApiKeyAuth": {
"in": "header",
"name": "X-API-KEY",
"type": "apiKey",
"description": "API key for authentication. You can obtain an API key from the Typecast API Console."
},
"BearerAuth": {
"type": "http",
"scheme": "bearer",
"bearerFormat": "JWT"
}
}
}
}
```