If you’ve ever struggled to find a high-quality, Indian-accented text-to-speech solution, Veena TTS by Maya Research is the game-changer you’ve been waiting for. Built on a powerful 3-billion parameter Llama-based architecture, Veena brings lifelike, expressive voices to both Hindi and English content, including seamless code-mixed scenarios. If you’re building voicebots, narrating audiobooks, or enhancing accessibility, Veena delivers natural speech output with ultra-low latency (sub-80ms on H100 GPUs) and clear 24kHz audio, thanks to its integration with the SNAC neural codec. What makes Veena stand out even more is its support for four distinct voices, Kavya, Agastya, Maitri, and Vinaya, each with its own vocal personality, giving you maximum flexibility for user experience design. Designed for production, it’s fully open source under the Apache 2.0 license and supports 4-bit quantization, making it both developer-friendly and deployment-ready. Simply put, this is not just a TTS model, it’s a complete voice generation engine purpose-built for India’s diverse linguistic landscape made for global use.
In this tutorial, you’ll learn how to install and run Veena TTS from start to end in minutes using transformers for free.
Prerequisites
The minimum system requirements for running this model are:
- GPU: 1x RTXA6000 or 1x A100
- Storage: 50 GB (preferable)
- VRAM: at least 16 GB
- Anaconda installed
Step-by-step process to install and run Veena TTS
For the purpose of this tutorial, we’ll use a GPU-powered Virtual Machine by NodeShift since it provides high compute Virtual Machines at a very affordable cost on a scale that meets GDPR, SOC2, and ISO27001 requirements. Also, it offers an intuitive and user-friendly interface, making it easier for beginners to get started with Cloud deployments. However, feel free to use any cloud provider of your choice and follow the same steps for the rest of the tutorial.
Step 1: Setting up a NodeShift Account
Visit app.nodeshift.com and create an account by filling in basic details, or continue signing up with your Google/GitHub account.
If you already have an account, login straight to your dashboard.
Step 2: Create a GPU Node
After accessing your account, you should see a dashboard (see image), now:
- Navigate to the menu on the left side.
- Click on the GPU Nodes option.
- Click on Start to start creating your very first GPU node.
These GPU nodes are GPU-powered virtual machines by NodeShift. These nodes are highly customizable and let you control different environmental configurations for GPUs ranging from H100s to A100s, CPUs, RAM, and storage, according to your needs.
Step 3: Selecting configuration for GPU (model, region, storage)
- For this tutorial, we’ll be using 1x A100 GPU, however, you can choose any GPU as per the prerequisites.
- Similarly, we’ll opt for 100GB storage by sliding the bar. You can also select the region where you want your GPU to reside from the available ones.
Step 4: Choose GPU Configuration and Authentication method
- After selecting your required configuration options, you’ll see the available GPU nodes in your region and according to (or very close to) your configuration. In our case, we’ll choose a 1x A100 80GB GPU node with 32vCPUs/131GB RAM/100GB SSD.
2. Next, you’ll need to select an authentication method. Two methods are available: Password and SSH Key. We recommend using SSH keys, as they are a more secure option. To create one, head over to our official documentation.
Step 5: Choose an Image
The final step is to choose an image for the VM, which in our case is Nvidia Cuda.
That’s it! You are now ready to deploy the node. Finalize the configuration summary, and if it looks good, click Create to deploy the node.
Step 6: Connect to active Compute Node using SSH
- As soon as you create the node, it will be deployed in a few seconds or a minute. Once deployed, you will see a status Running in green, meaning that our Compute node is ready to use!
- Once your GPU shows this status, navigate to the three dots on the right, click on Connect with SSH, and copy the SSH details that appear.
As you copy the details, follow the below steps to connect to the running GPU VM via SSH:
- Open your terminal, paste the SSH command, and run it.
2. In some cases, your terminal may take your consent before connecting. Enter ‘yes’.
3. A prompt will request a password. Type the SSH password, and you should be connected.
Output:
Next, If you want to check the GPU details, run the following command in the terminal:
!nvidia-smi
Step 7: Set up the project environment with dependencies
- Create a virtual environment using Anaconda.
conda create -n veena python=3.11 -y && conda activate veena
Output:
2. Once you’re inside the environment, install necessary dependencies to run the model.
pip install torch torchvision torchaudio einops timm pillow snac
pip install git+https://github.com/huggingface/transformers
pip install git+https://github.com/huggingface/accelerate
pip install sentencepiece bitsandbytes protobuf decord numpy
Output:
3. Install and run jupyter notebook.
conda install -c conda-forge --override-channels notebook -y
conda install -c conda-forge --override-channels ipywidgets -y
jupyter notebook --allow-root
4. If you’re on a remote machine (e.g., NodeShift GPU), you’ll need to do SSH port forwarding in order to access the jupyter notebook session on your local browser.
Run the following command in your local terminal after replacing:
<YOUR_SERVER_PORT> with the PORT allotted to your remote server (For the NodeShift server – you can find it in the deployed GPU details on the dashboard).
<PATH_TO_SSH_KEY> with the path to the location where your SSH key is stored.
<YOUR_SERVER_IP> with the IP address of your remote server.
ssh -L 8888:localhost:8888 -p <YOUR_SERVER_PORT> -i <PATH_TO_SSH_KEY> root@<YOUR_SERVER_IP>
Output:
After this copy the URL you received in your remote server:
And paste this on your local browser to access the Jupyter Notebook session.
Step 8: Download and Run the model
- Open a Python notebook inside Jupyter.
2. Download the model checkpoints.
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from snac import SNAC
import soundfile as sf
# Load model and tokenizer
model = AutoModelForCausalLM.from_pretrained(
"maya-research/veena-tts",
device_map="auto",
trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained("maya-research/veena-tts", trust_remote_code=True)
Output:
3. Initialize and load the model.
# Initialize SNAC decoder
snac_model = SNAC.from_pretrained("hubertsiuzdak/snac_24khz").eval().cuda()
# Control token IDs (fixed for Veena)
START_OF_SPEECH_TOKEN = 128257
END_OF_SPEECH_TOKEN = 128258
START_OF_HUMAN_TOKEN = 128259
END_OF_HUMAN_TOKEN = 128260
START_OF_AI_TOKEN = 128261
END_OF_AI_TOKEN = 128262
AUDIO_CODE_BASE_OFFSET = 128266
# Available speakers
speakers = ["kavya", "agastya", "maitri", "vinaya"]
def generate_speech(text, speaker="kavya", temperature=0.4, top_p=0.9):
"""Generate speech from text using specified speaker voice"""
# Prepare input with speaker token
prompt = f"<spk_{speaker}> {text}"
prompt_tokens = tokenizer.encode(prompt, add_special_tokens=False)
# Construct full sequence: [HUMAN] <spk_speaker> text [/HUMAN] [AI] [SPEECH]
input_tokens = [
START_OF_HUMAN_TOKEN,
*prompt_tokens,
END_OF_HUMAN_TOKEN,
START_OF_AI_TOKEN,
START_OF_SPEECH_TOKEN
]
input_ids = torch.tensor([input_tokens], device=model.device)
# Calculate max tokens based on text length
max_tokens = min(int(len(text) * 1.3) * 7 + 21, 700)
# Generate audio tokens
with torch.no_grad():
output = model.generate(
input_ids,
max_new_tokens=max_tokens,
do_sample=True,
temperature=temperature,
top_p=top_p,
repetition_penalty=1.05,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=[END_OF_SPEECH_TOKEN, END_OF_AI_TOKEN]
)
# Extract SNAC tokens
generated_ids = output[0][len(input_tokens):].tolist()
snac_tokens = [
token_id for token_id in generated_ids
if AUDIO_CODE_BASE_OFFSET <= token_id < (AUDIO_CODE_BASE_OFFSET + 7 * 4096)
]
if not snac_tokens:
raise ValueError("No audio tokens generated")
# Decode audio
audio = decode_snac_tokens(snac_tokens, snac_model)
return audio
def decode_snac_tokens(snac_tokens, snac_model):
"""De-interleave and decode SNAC tokens to audio"""
if not snac_tokens or len(snac_tokens) % 7 != 0:
return None
# De-interleave tokens into 3 hierarchical levels
codes_lvl = [[] for _ in range(3)]
llm_codebook_offsets = [AUDIO_CODE_BASE_OFFSET + i * 4096 for i in range(7)]
for i in range(0, len(snac_tokens), 7):
# Level 0: Coarse (1 token)
codes_lvl[0].append(snac_tokens[i] - llm_codebook_offsets[0])
# Level 1: Medium (2 tokens)
codes_lvl[1].append(snac_tokens[i+1] - llm_codebook_offsets[1])
codes_lvl[1].append(snac_tokens[i+4] - llm_codebook_offsets[4])
# Level 2: Fine (4 tokens)
codes_lvl[2].append(snac_tokens[i+2] - llm_codebook_offsets[2])
codes_lvl[2].append(snac_tokens[i+3] - llm_codebook_offsets[3])
codes_lvl[2].append(snac_tokens[i+5] - llm_codebook_offsets[5])
codes_lvl[2].append(snac_tokens[i+6] - llm_codebook_offsets[6])
# Convert to tensors for SNAC decoder
hierarchical_codes = []
for lvl_codes in codes_lvl:
device = next(snac_model.parameters()).device
tensor = torch.tensor(lvl_codes, dtype=torch.int32, device=device).unsqueeze(0)
if torch.any((tensor < 0) | (tensor > 4095)):
raise ValueError("Invalid SNAC token values")
hierarchical_codes.append(tensor)
# Decode with SNAC
with torch.no_grad():
audio_hat = snac_model.decode(hierarchical_codes)
return audio_hat.squeeze().clamp(-1, 1).cpu().numpy()
Output:
4. Run the model for inference.
# Hindi
text_hindi = "आज मैंने एक नई तकनीक के बारे में सीखा जो कृत्रिम बुद्धिमत्ता का उपयोग करके मानव जैसी आवाज़ उत्पन्न कर सकती है।"
audio = generate_speech(text_hindi, speaker="kavya")
sf.write("output_hindi_kavya.wav", audio, 24000)
# English
text_english = "Today I learned about a new technology that uses artificial intelligence to generate human-like voices."
audio = generate_speech(text_english, speaker="agastya")
sf.write("output_english_agastya.wav", audio, 24000)
# Code-mixed
text_mixed = "मैं तो पूरा presentation prepare कर चुकी हूं! कल रात को ही मैंने पूरा code base चेक किया।"
audio = generate_speech(text_mixed, speaker="maitri")
sf.write("output_mixed_maitri.wav", audio, 24000)
# English - Formal
text_english_formal = "Ladies and gentlemen, thank you for your attention and welcome to today's presentation."
audio = generate_speech(text_english_formal, speaker="vinaya")
sf.write("output_formal_vinaya.wav", audio, 24000)
Here’s the link to the output generated by the model for each text snippet shown above:
https://drive.google.com/drive/folders/1hgA101rfnCrP4sITbG2tlxUg9bn1z2Fe?usp=sharing
Conclusion
As we’ve seen, Veena TTS stands out as a production-grade, multilingual voice generation system built for the unique demands of Indian and/or global users, offering natural speech, low latency, and flexible voice options powered by a robust Llama-based architecture. Combined with the SNAC codec and Hugging Face’s transformer ecosystem, it makes high-quality TTS both accessible and developer-friendly. With NodeShift, deploying and running Veena becomes effortless, letting you skip the GPU setup hassle and go from installation to inference in minutes. If you’re experimenting or building at scale, NodeShift gives you the ideal cloud-native playground to unlock the full potential of Veena.
