API Reference

Text To Speech · HTTP Stream

POSThttps://api.shabd.tech/v1/text-to-speech/stream

Synthesizes text to audio and streams back binary chunks — each a 48-byte header followed by raw PCM-16 audio (see the field breakdown under the 200 response). Note: curl cannot frame the binary packet protocol — use Python or JavaScript.

Headers

x-api-keystringrequired
API Key for authentication
Example:
<your-api-key>
Content-Typestringrequired
Content type of the request body
Example:
application/json
x-request-idstringrequired
Unique request ID in UUID format
Example:
550e8400-e29b-41d4-a716-446655440000

Request bodyapplication/json

textstringrequired
Text to synthesize
Example:
Without language, we cannot communicate.
languagestringrequired
Target language code
Example:
english
genderstringrequired
Voice gender ("male" or "female")
Example:
male
sampleRateintegeroptional
Output audio sample rate in Hz
Example:
48000
speedstringoptional
Speed of speech (fast, normal, or slow). Default is normal if not specified.
Example:
normal
formatstringoptional
Output audio format (e.g. "wav")
Example:
wav
wordStreambooleanoptional
Whether to stream word-level timing information (optional)
Example:
false
returnTextbooleanoptional
Whether to return the input text alongside the audio (optional)
Example:
false
projectIdstringoptional
Optional project identifier
Example:
<your-project-id-here>

Responses

200Binary stream of audio chunks. Each chunk is a 48-byte header followed by raw PCM-16 audio; audio content begins at byte 48.

Field	Description	Size	Byte offset
requestId	Unique request id	36	0 – 35
status	0 = error, 1 = success	1	36
completed	0 = in progress, 1 = last chunk	1	37
sampleRate	Sample-rate indicator (kHz)	1	38
gender	0 = female, 1 = male	1	39
contentLength	Audio length in bytes (UInt32LE)	4	40 – 43
chunkId	Id of the audio chunk	1	44

On error, the chunk with status = 0 carries a UTF-8 error message as its content.

Body: application/octet-stream — binary

Example response

<binary audio — application/octet-stream>

400Bad Request
Bodyapplication/json
statusbooleanoptional
Example:
false
messagestringoptional
Example:
Invalid input parameters.
Example response
```
{  "status": false,  "message": "Invalid input parameters."}
```
401Unauthorized Access
Bodyapplication/json
statusbooleanoptional
Example:
false
messagestringoptional
Example:
Unauthorized access.
Example response
```
{  "status": false,  "message": "Unauthorized access."}
```
403Usage Limit Exceeded
Bodyapplication/json
statusbooleanoptional
Example:
false
messagestringoptional
Example:
Limit Exceeded
Example response
```
{  "status": false,  "message": "Limit Exceeded"}
```
500Internal server error
Bodyapplication/json
statusbooleanoptional
Example:
false
messagestringoptional
Example:
An unexpected error occurred. Please try again later.
Example response
```
{  "status": false,  "message": "An unexpected error occurred. Please try again later."}
```

Request

1# pip install httpx2import httpx3import asyncio4import wave5 6URL = "https://api.shabd.tech/v1/text-to-speech/stream"7HEADER_SIZE = 488SAMPLE_RATE = 480009headers = {10    "x-api-key": "<your-api-key>",11    "Content-Type": "application/json",12    "x-request-id": "<your-uuid-here>"13}14payload = {15    "text": "Without language, we cannot communicate.",16    "language": "english",17    "gender": "male",18    "sampleRate": SAMPLE_RATE,19    "speed": "normal"20}21 22async def main():23    buffer = bytearray()24    async with httpx.AsyncClient(timeout=60) as client:25        async with client.stream("POST", URL, json=payload, headers=headers) as response:26            if response.status_code != 200:27                print(f"HTTP Error: {response.status_code}")28                return29            async for chunk in response.aiter_bytes(chunk_size=512):30                buffer.extend(chunk)31                while True:32                    if len(buffer) < HEADER_SIZE:33                        break34                    content_length = int.from_bytes(buffer[40:44], "little")35                    frame_size = HEADER_SIZE + content_length36                    if len(buffer) < frame_size:37                        break38                    frame = buffer[:frame_size]39                    buffer = buffer[frame_size:]40                    header = frame[:HEADER_SIZE]41                    audio = frame[HEADER_SIZE:frame_size]42                    status = header[36]43                    completed = header[37]44                    chunk_id = header[44]45                    if not status:46                        print("Server Error:", audio.decode("utf-8", "ignore"))47                        return48                    with wave.open(f"SynthesizedAudio_Chunk{chunk_id}.wav", "wb") as wf:49                        wf.setnchannels(1)50                        wf.setsampwidth(2)51                        wf.setframerate(SAMPLE_RATE)52                        wf.writeframes(audio)53                    print(f"Saved chunk {chunk_id}: {len(audio)} bytes")54                    if completed == 1:55                        print("Audio synthesis completed")56                        return57 58asyncio.run(main())

1// npm install node-fetch wav2const fetch = require("node-fetch");3const wav = require("wav");4 5const url = "https://api.shabd.tech/v1/text-to-speech/stream";6const sampleRate = 48000;7 8const response = await fetch(url, {9    method: "POST",10    headers: {11        "x-api-key": "<your-api-key>",12        "Content-Type": "application/json",13        "x-request-id": "<your-uuid-here>"14    },15    body: JSON.stringify({16        text: "Without language, we cannot communicate.",17        language: "english",18        gender: "male",19        sampleRate,20        speed: "normal"21    })22});23 24let buffer = Buffer.alloc(0);25const HEADER_SIZE = 48;26for await (const value of response.body) {27    buffer = Buffer.concat([buffer, value]);28    while (true) {29        if (buffer.length < HEADER_SIZE) break;30        const contentLength = buffer.readUInt32LE(40);31        const frameSize = HEADER_SIZE + contentLength;32        if (buffer.length < frameSize) break;33        const header = buffer.slice(0, HEADER_SIZE);34        const audio = buffer.slice(HEADER_SIZE, frameSize);35        buffer = buffer.slice(frameSize);36        const status = header[36];37        const completed = header[37];38        const chunkId = header[44];39        if (!status) { console.error("Error:", audio.toString()); break; }40        const writer = new wav.Writer({ channels: 1, sampleRate, bitDepth: 16 });41        writer.pipe(fs.createWriteStream(`SynthesizedAudio_Chunk${chunkId}.wav`));42        writer.write(audio);43        writer.end();44        console.log(`Saved chunk ${chunkId}: ${audio.length} bytes`);45        if (completed) { console.log("Audio synthesis completed"); break; }46    }47}

Text To Speech · HTTP Stream

Headers

Request bodyapplication/json

Responses

Bodyapplication/json

Bodyapplication/json

Bodyapplication/json

Bodyapplication/json