> For clean Markdown of any page, append `.md` to the page URL. > For a complete documentation index, see https://docs.sarvam.ai/llms.txt. > For full documentation content in one file, see https://docs.sarvam.ai/llms-full.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server. # Text-to-Speech Rest API > Real-time conversion of text into speech using customizable voices. Instant audio generation with multiple voice options and various audio formats for Indian languages. Provides a synchronous REST endpoint where a POST request with text returns base64-encoded audio as response. The JSON response contains an `audios` array of **base64-encoded WAV strings**, not raw binary. Decode before saving or playing: ```python import base64 combined = "".join(response.audios) wav_bytes = base64.b64decode(combined) with open("output.wav", "wb") as f: f.write(wav_bytes) ``` See [TTS best practices](/api-reference-docs/api-guides-tutorials/text-to-speech/best-practices) for JavaScript and streaming examples. **Common use cases:** * **Story narration** — Generate expressive audio for audiobooks and narratives * **Podcast generation** — Create natural-sounding voiceovers for episodes at scale * **Content creation** — Add voice to blogs, articles, and social media posts * **E-learning** — Build multilingual course material with clear pronunciation ## What You Can Do Pick from male and female speakers — each with distinct tone and style.
Pass the `speaker` param to switch instantly. Hindi, Bengali, Tamil, Telugu, Kannada, Malayalam, Marathi, Gujarati, Punjabi, Odia, and English (Indian accent).
Set via `target_language_code`. Send long-form text in a single request (v3). No need to chunk or paginate your input. Speed up or slow down speech with the `pace` parameter — range `0.5` to `2.0` for v3. 8kHz to 48kHz output. Higher rates (32k, 44.1k, 48k) available in **v3 REST API only**. Default: 24kHz. Response is base64-encoded. Supports WAV, MP3, Linear16, Mulaw, Alaw, Opus, FLAC, and AAC. ## Model: Bulbul v3 Bulbul v3 is purpose-built for Indian languages and accents. It handles code-mixed text (e.g., Hinglish), number normalization, and natural prosody out of the box — with minimal preprocessing needed. ## Text to Speech Features

Basic Text to Speech Synthesis

Convert text to natural-sounding speech with high quality. Features include:

```python from sarvamai import SarvamAI from sarvamai.play import save client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY") # Convert text to speech audio = client.text_to_speech.convert( target_language_code="en-IN", text="Welcome to Sarvam AI!", model="bulbul:v3", speaker="shubh" ) save(audio, "output1.wav") ``` ```javascript import { SarvamAIClient } from "sarvamai"; import fs from "fs"; const client = new SarvamAIClient({ apiSubscriptionKey: "YOUR_SARVAM_API_KEY" }); const response = await client.textToSpeech.convert({ text: "Welcome to Sarvam AI!", model: "bulbul:v3", speaker: "shubh", target_language_code: "en-IN" }); const audio = Buffer.from(response.audios.join(""), "base64"); fs.writeFileSync("output.wav", audio); ``` ```bash curl -X POST https://api.sarvam.ai/text-to-speech \ -H "api-subscription-key: " \ -H "Content-Type: application/json" \ -d '{ "text": "Welcome to Sarvam AI!", "target_language_code": "en-IN", "speaker": "shubh", "model": "bulbul:v3" }' ```

Available Voices

Choose from 30+ natural-sounding voices for different use cases and languages.

**Male:** Shubh (default), Aditya, Rahul, Rohan, Amit, Dev, Ratan, Varun, Manan, Sumit, Kabir, Aayan, Ashutosh, Advait, Anand, Tarun, Sunny, Mani, Gokul, Vijay, Mohit, Rehan, Soham **Female:** Ritu, Priya, Neha, Pooja, Simran, Kavya, Ishita, Shreya, Roopa, Tanya, Shruti, Suhani, Kavitha, Rupali **Female:** Anushka (default), Manisha, Vidya, Arya **Male:** Abhilash, Karun, Hitesh ```python import base64 from sarvamai import SarvamAI client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY") response = client.text_to_speech.convert( text="Welcome to Sarvam AI!", model="bulbul:v3", target_language_code="en-IN", speaker="shubh" ) with open("output.wav", "wb") as f: f.write(base64.b64decode("".join(response.audios))) ``` ```javascript import { SarvamAIClient } from "sarvamai"; import fs from "fs"; const client = new SarvamAIClient({ apiSubscriptionKey: "YOUR_SARVAM_API_KEY" }); const response = await client.textToSpeech.convert({ text: "Hello, how are you?", target_language_code: "hi-IN", model: "bulbul:v3", speaker: "shubh" }); const audio = Buffer.from(response.audios.join(""), "base64"); fs.writeFileSync("output.wav", audio); ``` ```bash # Generate speech with Shubh's voice curl -X POST https://api.sarvam.ai/text-to-speech \ -H "api-subscription-key: " \ -H "Content-Type: application/json" \ -d '{ "text": "Welcome to Sarvam AI!", "target_language_code": "en-IN", "speaker": "shubh", "model": "bulbul:v3" }' # Generate speech with Priya's voice curl -X POST https://api.sarvam.ai/text-to-speech \ -H "api-subscription-key: " \ -H "Content-Type: application/json" \ -d '{ "text": "Welcome to Sarvam AI!", "target_language_code": "en-IN", "speaker": "priya", "model": "bulbul:v3" }' # Generate speech with Roopa's voice curl -X POST https://api.sarvam.ai/text-to-speech \ -H "api-subscription-key: " \ -H "Content-Type: application/json" \ -d '{ "text": "Welcome to Sarvam AI!", "target_language_code": "en-IN", "speaker": "roopa", "model": "bulbul:v3" }' ```

Speech Customization

Fine-tune the speech output with various parameters:

  • Adjust speech rate (pace)
  • Configure audio quality (sample rate)
```python import base64 from sarvamai import SarvamAI client = SarvamAI( api_subscription_key="YOUR_SARVAM_API_KEY" ) audio = client.text_to_speech.convert( text="Welcome to Sarvam AI!", model="bulbul:v3", target_language_code="en-IN", speaker="shubh", pace=1.2, speech_sample_rate=24000 ) combined_audio = "".join(audio.audios) b64_file = base64.b64decode(combined_audio) with open("output1.wav", "wb") as f: f.write(b64_file) ``` ```javascript import { SarvamAIClient } from "sarvamai"; import fs from "fs"; const client = new SarvamAIClient({ apiSubscriptionKey: "YOUR_SARVAM_API_KEY" }); const response = await client.textToSpeech.convert({ text: "Welcome to Sarvam AI!", model: "bulbul:v3", target_language_code: "en-IN", speaker: "shubh", pace: 1.2, speech_sample_rate: 24000 }); const audio = Buffer.from(response.audios.join(""), "base64"); fs.writeFileSync("output.wav", audio); ``` ```bash curl -X POST https://api.sarvam.ai/text-to-speech \ -H "api-subscription-key: " \ -H "Content-Type: application/json" \ -d '{ "text": "Welcome to Sarvam AI!", "model": "bulbul:v3", "speaker": "shubh", "pace": 1.2, "target_language_code": "en-IN", "speech_sample_rate": 24000 }' ``` ## API Response Format | Field | Type | Description | | ------------ | ------ | --------------------------------------------------------------------- | | `request_id` | string | Unique identifier for the request | | `audios` | array | Base64-encoded audio files. Each element corresponds to an input text | **Supported audio formats:** WAV (default), MP3, Linear16, Mulaw, Alaw, Opus, FLAC, AAC ```json { "request_id": "20241115_12345678-1234-5678-1234-567812345678", "audios": [ "UklGRiQAAABXQVZFZm10IBAAAAABAAEAQB8AAAB9AAACABAAZGF0YQAAAAA..." ] } ``` **Python:** ```python import base64 audio_base64 = response.audios[0] audio_bytes = base64.b64decode(audio_base64) with open("output.wav", "wb") as f: f.write(audio_bytes) ``` **JavaScript:** ```javascript import fs from "fs"; const audioBase64 = response.audios[0]; const audioBuffer = Buffer.from(audioBase64, 'base64'); fs.writeFileSync('output.wav', audioBuffer); ``` ## Error Responses All errors return a JSON object with an `error` field containing details about what went wrong. ### Error Response Structure ```json { "error": { "message": "Human-readable error description", "code": "error_code_for_programmatic_handling", "request_id": "unique_request_identifier" } } ``` ### Error Codes Reference | HTTP Status | Error Code | When This Happens | What To Do | | ----------- | ---------------------------- | ------------------------------------------------ | -------------------------------------------------- | | `400` | `invalid_request_error` | Missing required parameters or malformed request | Check `text` and `target_language_code` fields | | `403` | `invalid_api_key_error` | API key is invalid, missing, or expired | Verify your API key in the dashboard | | `422` | `unprocessable_entity_error` | Text too long or invalid speaker/model | Keep text under 1500 chars (v2) or 2500 chars (v3) | | `429` | `insufficient_quota_error` | API quota or rate limit exceeded | Wait for reset or upgrade your plan | | `500` | `internal_server_error` | Unexpected server error | Retry the request; contact support if persistent | ### Example Error Response ```json { "error": { "message": "Text exceeds maximum length of 1500 characters for bulbul:v3", "code": "unprocessable_entity_error", "request_id": "20241115_abc12345" } } ``` ```python from sarvamai import SarvamAI from sarvamai.core.api_error import ApiError client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY") try: response = client.text_to_speech.convert( text="Welcome to Sarvam AI!", target_language_code="en-IN", speaker="shubh", model="bulbul:v3" ) # Process audio... except ApiError as e: if e.status_code == 400: print(f"Bad request: {e.body}") elif e.status_code == 403: print("Invalid API key. Check your credentials.") elif e.status_code == 422: print(f"Invalid parameters: {e.body}") elif e.status_code == 429: print("Rate limit exceeded. Wait and retry.") else: print(f"Error {e.status_code}: {e.body}") ``` Check out our detailed [API Reference](/api-reference-docs/text-to-speech/convert) to explore Text to Speech and all available options. Need help? Contact us on [discord](https://discord.com/invite/5rAsykttcs) for guidance.