3b:I[603186,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"PageMetadata"] 3c:I[831915,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"SearchResultsImage"] 3d:I[430295,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],""] 3e:I[299632,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"default"] 3f:I[400402,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"EndpointCTA"] 40:I[331699,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"ModelContent"] 47:I[647007,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"default"] 48:I[946424,["/_next/static/chunks/0410xx54a0zpq.js","/_next/static/chunks/3pd35vh8xvq70.js","/_next/static/chunks/10s2t_rqdffvv.js","/_next/static/chunks/3yq_f6lmn4_qv.js","/_next/static/chunks/1la1pzs9j8idc.js","/_next/static/chunks/24mqrfe34h1_g.js","/_next/static/chunks/3-axx3uk1c0-h.js","/_next/static/chunks/14bn3mp7hfp6x.js","/_next/static/chunks/1_xj2c5utin8v.js","/_next/static/chunks/2tdv_pqf9hlyu.js","/_next/static/chunks/44n8gpf9x1nn2.js","/_next/static/chunks/1e7h6z92_w2-c.js","/_next/static/chunks/1i7ii0s7yr3co.js","/_next/static/chunks/1geiruffics9x.js","/_next/static/chunks/1lfygqrocy8iw.js","/_next/static/chunks/0q47y720ck5-9.js","/_next/static/chunks/39sdfeeqouoq5.js","/_next/static/chunks/3mi_myijg06x1.js","/_next/static/chunks/2_c926k09c1th.js","/_next/static/chunks/0bvim9_g0hu68.js","/_next/static/chunks/3xav1ddquf5gq.js","/_next/static/chunks/401qc4n4ul0k5.js","/_next/static/chunks/34kl4tv6yifcg.js","/_next/static/chunks/0nm6za7z70maq.js","/_next/static/chunks/44e22u2t839t0.js","/_next/static/chunks/0czpc-f8wjri8.js","/_next/static/chunks/2aq42x9vabhfb.js","/_next/static/chunks/24_gg9g_4g22i.js","/_next/static/chunks/00yjb-caoi49j.js","/_next/static/chunks/1l0z2zsen9vs2.js","/_next/static/chunks/1p76x0jjzflrw.js","/_next/static/chunks/2d25fo_l-7jx3.js","/_next/static/chunks/20n58999dvhts.js","/_next/static/chunks/1yd7p-515l2jc.js","/_next/static/chunks/2pqhypz2vo8mu.js","/_next/static/chunks/3udsmcp2swosl.js","/_next/static/chunks/2cw7k2bes-1su.js","/_next/static/chunks/295ffwyuqa4b6.js"],"default"] 38:["$","$L3b",null,{"hasPartnerEndpoint":false}] 41:T2302,# Model Overview ### Description: Studio Voice is a speech enhancement model from NVIDIA. Studio Voice enhances input speech recorded with low-quality microphones in noisy or reverberant environments, producing studio-quality speech. This model is ready for commercial use. ### License/Terms of Use: Use of this model is governed by the [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/). ### Deployment Geography: Global ### Use Case: Studio Voice models removes microphone artifacts and room reverberations to produce studio quality voice. It is intended to be used by content developers and broadcasters. ### Release Date: NGC [03/16/2026] via [afx_studio_voice](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/maxine/models/afx_studio_voice) ## Model Architecture **Architecture Type:** Convolution Neural Networks (CNNs), Transformers, Generative Adversarial Networks (GANs) **Network Architecture:** Encoder-Decoder **Number of model parameters:** 183M ## Input(s): **Input Type(s):** Audio **Input Format(s):** PCM F32 **Input Parameters:** One-Dimensional (1D) **Other Properties Related to Input:** Pulse Code Modulation (PCM) audio samples with no encoding or pre-processing; 16kHz or 48kHz sampling rate required. ## Output(s): **Output Type(s):** Audio **Output Format:** PCM F32 **Output Parameters:** One-Dimensional (1D) **Other Properties Related to Output:** PCM audio samples at input sampling rate with no encoding or post-processing. Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions. ## Software Integration **Runtime Engine(s):** * Audio Effects 2.1.0 **Supported Hardware Microarchitecture Compatibility:** * NVIDIA Ampere * NVIDIA Blackwell * NVIDIA Hopper * NVIDIA Lovelace * NVIDIA Turing **Preferred/Supported Operating System(s):** * Linux * Windows The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment. This AI model can be embedded as an Application Programming Interface (API) call into the software environment described above. ## Model Version(s) StudioVoice v0.5.1 ## Training, Testing, and Evaluation Datasets **Data Modality:** * Audio **Audio Training Data Size:** * Less than 10,000 Hours **Dataset partition:** Training [80%], Testing [10%], Validation [10%] NVIDIA models are trained on a diverse set of public and proprietary datasets. The Studio Voice model is trained on a dataset that comprises of diverse English accents and different types of microphone devices. **Data Collection Method by dataset:** [Hybrid: Human, Synthetic] **Labeling Method by dataset:** [Hybrid: Human, Synthetic] **Link:** [DAPS](https://ccrma.stanford.edu/~gautham/Site/daps.html) **Properties:** The DAPS dataset has 15 versions of audio (3 professional versions and 12 consumer device/real-world environment combinations). Each version consists of about 4.5 hours of data (about 14 minutes from each of 20 speakers). **Link:** [LibriTTS](https://www.openslr.org/60/) **Properties:** LibriTTS is a multi-speaker English corpus of approximately 585 hours of read English speech, which is resampled at 16kHZ. **Link:** [VCTK](https://datashare.ed.ac.uk/handle/10283/3443) **Properties:** This CSTR VCTK Corpus includes speech data uttered by 110 English speakers with various accents. Each speaker reads out about 400 sentences, which were selected from a newspaper, the rainbow passage and an elicitation paragraph used for the speech accent archive. **Link:** [HiFi-TTS](https://www.openslr.org/109/) **Properties:** A multi-speaker English dataset for training text-to-speech models. The HiFi-TTS dataset contains about 291.6 hours of speech from 10 speakers with at least 17 hours per speaker sampled at 44.1 kHz. **Link:** [Device Recorded VCTK (DR-VCTK)](https://github.com/nii-yamagishilab/downloader-DR-VCTK-complete) **Properties:** Device recorded version of VCTK dataset on common consumer devices (laptop, tablet and smartphone) in office environment. This dataset contains 109 English speakers with different accents. There are around 400 sentences available from each speaker. For this recording, 8 different microphones were used. This dataset contains around 250 Gb of re-recorded speech. **Link:** [Dataset of impulse responses from variable acoustics room Arni at Aalto Acoustic Labs](https://zenodo.org/records/6985104#.Y7_vv3ZBy3C) **Properties:** A dataset of impulse responses collected in the variable acoustics laboratory Arni at Acoustics Lab of Aalto University, Espoo, Finland. IRs of 5342 configurations of sound absorption in Arni are included in the dataset. Each of them were measured using an omnidirectional sound source and 5 sound receivers. For each configuration, 5 impulse reponses (IRs) were captured. The total number of measurements in the dataset is 132 037. **Link:** [Room Impulse Response and Noise Database](https://www.openslr.org/28/) **Properties:** A database of simulated and real room impulse responses, isotropic and point-source noises. The audio files in this data are all in 16KHz sampling rate and 16-bit precision. **Link:** [DNS Challenge 5](https://github.com/microsoft/DNS-Challenge/tree/2db96d5f75257df764a6ef66513b4b97bc707f30) **Properties:** Collated dataset of clean speech, noise and impulse response provided by Microsoft for the ICASSP 2023 Deep Noise Suppression Challenge. **Link:** [AudioSet](https://research.google.com/audioset/download.html) **Properties:** AudioSet consists of an expanding ontology of 632 audio event classes and a collection of 2,084,320 human-labeled 10-second sound clips drawn from publicly available internet scale data. **Link:** Multi-Language and Emotions Speech Dataset **Properties:** Multi-Language and Emotions Speech Dataset contains high quality speech data worth approximately 140 hours and contains approximately 80 unique speakers. The dataset contains different english accents and contains different emotions as well. This dataset is taken from publicly available internet scale data. **Link:** Audio2Gesture Bones Dataset **Properties:** Audio2Gesture Bones Dataset is a speech dataset which is bought from Bones.studio company. It contains only 2 speakers covering all different types of emotions. It is a small dataset of around 4 GB and contains 7 hours of speech data. ### Testing Datasets **Data Collection Method by dataset:** [Hybrid: Human, Synthetic] **Labeling Method by dataset:** [Hybrid: Human, Synthetic] **Properties:** The Studio Voice model is tested on a dataset that comprises of diverse English accents and different types of microphone devices. Test data is taken by sampling 10% of the training dataset mentioned above. The modality and data type is same as that of the training dataset. ### Evaluation Datasets **Data Collection Method by dataset:** [Hybrid: Human, Synthetic] **Labeling Method by dataset:** [Hybrid: Human, Synthetic] **Properties:** The Studio Voice model is evaluated on a dataset that comprises of diverse English accents and different types of microphone devices. Evaluation data is taken by sampling 10% of the training dataset mentioned above. The modality and data type is same as that of the training dataset. ## Inference **Acceleration Engine:** Tensor(RT), [Triton](https://developer.nvidia.com/triton-inference-server) **Test Hardware:** * T4, T10, A30, A100, A2, A10, A16, A40, L4, L40, H100, B40, B100 * RTX 4080, RTX 4090, RTX 5070, RTX 5080, RTX 5090 ## Ethical Considerations NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. For more detailed information on ethical considerations for this model, please see the Model Card++ Bias, Explainability, Safety & Security, and Privacy Subcards. Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).42:T58a,| Field | Response | |-------|----------| | Intended Task/Domain: | Voice Enhancement | | Model Type: | Speech Enhancement | | Intended Users: | Content developers and Broadcasters | | Output: | Audio | | Describe how the model works: | This model enhances the voice in input audio by removing acoustic noise, room reverberation, and bad frequency characteristics of the recording device while maintaining the linguistic and vocal properties. | | Name the adversely impacted groups this has been tested to deliver comparable outcomes regardless of: | Age (18+), Gender | | Verified to have met prescribed NVIDIA quality standards: | Yes | | Performance Metrics: | Quality Mean Opinion Score (MOS), Non-Intrusive Speech Quality Assessment (NISQA), Latency, Throughput. | | Technical Limitations & Mitigation: | The model may not work well on a variety of demographic and regional representations of English or with very noisy or very low quality inputs. Exclamatory emotions may not be transferred to the output. | | Potential Known Risks: | The model may not preserve some speaker characteristics like pitch/fundamental frequency of the speakers voice and might also impact intelligibility of speech. | | Licensing: | [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) |43:T512,| Field | Response | |-------|----------| | Generatable or reverse engineerable personal data? | No | | Personal data used to create this model? | Yes | | Was consent obtained for any personal data used? | Yes | | How often is dataset reviewed? | Before Release | | Is a mechanism in place to honor data subject right of access or deletion of personal data? | Yes | | If personal data collected for the development of the model, was it collected directly by NVIDIA? | No | | If personal data collected for the development of the model by NVIDIA, do you maintain or have access to disclosures made to data subjects? | NO | | If personal data collected for the development of this AI model, was it minimized to only what was required? | Yes | | Was data from user interactions with the AI model (e.g. user input and prompts) used to train the model? | No | | Is there provenance for all datasets used in training? | Yes | | Does data labeling (annotation, metadata) comply with privacy laws? | Yes | | Is data compliant with data subject requests for data correction or removal, if such a request was made? | No, not possible with externally-sourced data. | | Applicable Privacy Policy | https://www.nvidia.com/en-us/about-nvidia/privacy-policy/ |44:T8d6,## Getting Started NVIDIA Studio Voice NIM uses gRPC APIs for inferencing requests. Following instructions demonstrate the usage of Studio Voice NIM model using Python client. ### Prerequisites You will need a system with `git` and `Python 3.10+` installed. ### Setup NVIDIA Studio Voice Python client Download the Studio Voice Python client code by cloning the [NIM gRPC Client Repository](https://github.com/NVIDIA-Maxine/nim-clients): ```bash git clone https://github.com/NVIDIA-Maxine/nim-clients.git cd nim-clients/studio-voice ``` Install the dependencies for the NVIDIA Studio Voice Python client: ```bash sudo apt-get install python3-pip pip install -r requirements.txt ``` ### Run Python Client Navigate to the scripts directory. ```bash cd scripts ``` Send the gRPC requests ```bash python studio_voice.py --preview-mode \ --ssl-mode TLS \ --target grpc.nvcf.nvidia.com:443 \ --function-id <%- nvcfFunctionId %> \ --api-key <%- apiKey %> \ --input \ --output ``` Note the requirements for input file: - The supported format is 16-bit mono channel wav file. - The size limit for input file is 32 MB. - The duration limit for input file is 6 min. Command line arguments: - `--preview-mode` - Flag to send request to preview NVCF server on https://build.nvidia.com/nvidia/studiovoice/api. - `--ssl-mode` - Set the SSL mode to TLS or MTLS. Defaults to no SSL. When running preview, TLS mode must be used with default root certificate. - `--target ` - URI of NIM's gRPC service. Use grpc.nvcf.nvidia.com:443 when hosted on NVCF. (Default: `127.0.0.1:8001`) - `--api-key <%- apiKey %>` - NGC API key required for authentication. Utilized when using `TRY API` ignored otherwise. - `--function-id ` - Function ID for the feature. - `--input ` - The path to the input audio file. (Default: `../assets/studio_voice_48k_input.wav`) - `--output ` - The path to the output audio file. (Default: `./studio_voice_48k_output.wav`) - `--streaming` - Flag to enable grpc streaming mode. Refer the [Studio Voice NIM documentation](https://docs.nvidia.com/nim/maxine/studio-voice/latest/index.html) for more information. 45:T177e,## Step 1. Generate API Key ::generate-api-key ## Step 2. Pull and Run the NIM ```bash $ docker login nvcr.io Username: $oauthtoken Password: ``` NVIDIA Studio Voice NIM uses gRPC APIs for inferencing requests. A NGC API KEY is required to download the appropriate models and resources when starting the NIM. If you are not familiar with how to create the `NGC_API_KEY` environment variable, the simplest way is to export it in your terminal: ```bash export NGC_API_KEY= ``` Run one of the following commands to make the key available at startup: ```bash # If using bash echo "export NGC_API_KEY=" >> ~/.bashrc # If using zsh echo "export NGC_API_KEY=" >> ~/.zshrc ``` Other, more secure options include saving the value in a file, so that you can retrieve with `cat $NGC_API_KEY_FILE`, or using a [password manager](https://www.passwordstore.org/). The following commands launch the Studio Voice NIM container with the gRPC service. Find reference to runtime parameters for the container [here](https://docs.nvidia.com/nim/maxine/studio-voice/latest/getting-started.html#runtime-parameters-for-the-container). Ensure you use the appropriate `NIM_MODEL_PROFILE` for your GPU. For more information about `NIM_MODEL_PROFILE`, refer to the the [NIM Model Profile Table](https://docs.nvidia.com/nim/maxine/studio-voice/latest/model-profile-table.html). Please note, the flag --gpus all is used to assign all available GPUs to the docker container. This fails on multiple GPU unless all GPUs are same. To assign specific GPU to the docker container (in case of different multiple GPUs available in your machine) use --gpus '"device=0,1,2..."' ### Streaming Mode ```bash docker run -it --rm --name=studio-voice-nim \ --runtime=nvidia \ --gpus all \ --shm-size=8GB \ -e NGC_API_KEY=$NGC_API_KEY \ -e NIM_MODEL_PROFILE= \ -e STREAMING=true \ -p 8000:8000 \ -p 8001:8001 \ nvcr.io/nim/nvidia/studio-voice:latest ``` ### Transactional Mode ```bash docker run -it --rm --name=studio-voice-nim \ --runtime=nvidia \ --gpus all \ --shm-size=8GB \ -e NGC_API_KEY=$NGC_API_KEY \ -e NIM_MODEL_PROFILE= \ -e FILE_SIZE_LIMIT=36700160 \ -e STREAMING=false \ -p 8000:8000 \ -p 8001:8001 \ nvcr.io/nim/nvidia/studio-voice:latest ``` If the command runs successfully, you will get an output ending similar to the following: ```bash I1126 09:22:21.048202 31 grpc_server.cc:2558] "Started GRPCInferenceService at 127.0.0.1:9001" I1126 09:22:21.048377 31 http_server.cc:4704] "Started HTTPService at 127.0.0.1:9000" I1126 09:22:21.089295 31 http_server.cc:362] "Started Metrics Service at 127.0.0.1:9002" [INFO AI4M BASE LOGGER 2026-03-06 06:54:09.052 PID:235] Listening to 0.0.0.0:8001 ``` By default Studio Voice gRPC service is hosted on port `8001`. You will have to use this port for inferencing requests. ## Step 3. Test the NIM We have provided a sample client script file in our GitHub repo. The script could be used to invoke the Docker container using the following instructions. Download the Studio Voice Python client code by cloning the [NIM gRPC Client Repository](https://github.com/NVIDIA-Maxine/nim-clients): ```bash git clone https://github.com/NVIDIA-Maxine/nim-clients.git cd nim-clients/studio-voice ``` Install the dependencies for the NVIDIA Studio Voice Python client: ```bash sudo apt-get install python3-pip pip install -r requirements.txt ``` Go to scripts directory ```bash cd scripts ``` Ensure the client mode matches the NIM's `STREAMING` mode set in Step 2. Run the command to send gRPC request (transactional mode i.e. STREAMING=false): ```bash python studio_voice.py --target --input --output ``` For streaming mode: ```bash python studio_voice.py --target --input --output --streaming --model-type 48k-hq ``` When using `--streaming` mode, ensure the selected `--model-type` (48k-hq, 48k-ll, or 16k-hq) aligns with the `NIM_MODEL_PROFILE` Model Type configuration to maintain compatibility . To view details of command line arguments run this command ```bash python studio_voice.py -h ``` You will get a response similar to the following. ```bash usage: studio_voice.py [-h] [--ssl-mode {MTLS,TLS}] [--ssl-key SSL_KEY] [--ssl-cert SSL_CERT] [--ssl-root-cert SSL_ROOT_CERT] [--target TARGET] [--input INPUT] [--output OUTPUT] [--api-key API_KEY] [--function-id FUNCTION_ID] [--streaming] [--model-type {48k-hq,48k-ll,16k-hq}] Process wav audio files using gRPC and apply studio-voice. options: -h, --help show this help message and exit --preview-mode Flag to send request to preview NVCF server on https://build.nvidia.com/nvidia/studiovoice/api. --ssl-mode {MTLS,TLS} Flag to set SSL mode, default is None --ssl-key SSL_KEY The path to ssl private key. --ssl-cert SSL_CERT The path to ssl certificate chain. --ssl-root-cert SSL_ROOT_CERT The path to ssl root certificate. --target TARGET IP:port of gRPC service, when hosted locally. Use grpc.nvcf.nvidia.com:443 when hosted on NVCF. --input INPUT The path to the input audio file. --output OUTPUT The path for the output audio file. --api-key API_KEY NGC API key required for authentication, utilized when using TRY API ignored otherwise --function-id FUNCTION_ID NVCF function ID for the service, utilized when using TRY API ignored otherwise --streaming Flag to enable grpc streaming mode. --model-type {48k-hq,48k-ll,16k-hq} Studio Voice model type, default is 48k-hq. ``` For more details on getting started with this NIM including configuring using parameters, visit the [NVIDIA Studio Voice NIM Docs](https://docs.nvidia.com/nim/maxine/studio-voice/latest/index.html). 46:T14ca,## Step 1. Apply for Access This NIM is available through [Apply to Self-Host](https://www.nvidia.com/en-us/ai/nim-notifyme/). Access gives you an NGC API key with access to the Studio Voice images, Helm charts, and models. ## Step 2. Prerequisites Before deployment, make sure your environment is prepared for Holoscan for Media and/or SMPTE ST 2110: - Kubernetes cluster with a GPU node (L40S or NVIDIA RTX PRO 6000 Blackwell Server Edition). - Rivermax license and runtime support — see [Rivermax License](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/latest/getting-started.html#rivermax-license). - High-speed network attachment (`NetworkAttachmentDefinition`) for ST 2110 traffic. - NGC API key and Kubernetes secrets for image pull and model download. For full prerequisite details, refer to the [Getting Started](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/latest/getting-started.html) guide. Note the name of your `NetworkAttachmentDefinition` and your GPU node's hostname — you will need both in Step 4. ```bash kubectl get network-attachment-definitions kubectl get nodes -o wide ``` ## Step 3. Pull the Helm Chart Authenticate and pull the end-to-end demo Helm chart: ```bash export NGC_API_KEY= helm repo add nvidia-nim \ https://helm.ngc.nvidia.com/nim/nvidia/ \ --username='$oauthtoken' \ --password=$NGC_API_KEY helm repo update helm pull nvidia-nim/nvidia-studio-voice-h4m-sample --version 1.2.0 ``` ## Step 4. Install the ST 2110 Static Mode Pipeline This deployment installs the complete pipeline — sender, Studio Voice NIM service, and receiver. The receiver exposes an SRT stream that can be previewed in VLC, ffplay, or other SRT players. This sample uses **static ST 2110 mode**, where multicast addresses and ports are predefined in `values-st2110.yaml` unless overridden during installation. --- ### Minimal Installation Using Global Configuration If all services use the same GPU node, network attachment, and image pull secret, configure them globally instead of per component: ```bash helm upgrade --install studio-voice-h4m-sample \ nvidia-studio-voice-h4m-sample-1.2.0.tgz \ -f <(tar -xOf nvidia-studio-voice-h4m-sample-1.2.0.tgz \ nvidia-studio-voice-h4m-sample/values-st2110.yaml) \ --set global.nodeSelector.hostname= \ --set global.image.secret= \ --set global.network.name= \ --set nvidia-studio-voice-h4m-service.ngc.secretName= ``` This applies the same node, network, and image pull secret configuration across the sender, receiver, and Studio Voice NIM service. --- ### Optional: Install Using Per-Component Configuration Use this option when components require different GPU nodes, network attachments, or image pull secrets. ```bash helm upgrade --install studio-voice-h4m-sample \ nvidia-studio-voice-h4m-sample-1.2.0.tgz \ -f <(tar -xOf nvidia-studio-voice-h4m-sample-1.2.0.tgz \ nvidia-studio-voice-h4m-sample/values-st2110.yaml) \ --set sender.nodeSelector.hostname= \ --set receiver.nodeSelector.hostname= \ --set nvidia-studio-voice-h4m-service.nodeSelector.hostname= \ --set nvidia-studio-voice-h4m-service.ngc.secretName= ``` --- ### Optional: Configure Network Attachment Per Component Only required if not using the default `media-a-tx-net` network attachment. Append the following to the install command: ```bash --set sender.network.name= \ --set receiver.network.name= \ --set nvidia-studio-voice-h4m-service.network.name= ``` --- ### Optional: Configure Image Pull Secret Per Component Only required if not using the global image pull secret or default `ngc-image-secret`. Append the following to the install command: ```bash --set sender.image.secret= \ --set receiver.image.secret= \ --set nvidia-studio-voice-h4m-service.image.secret= ``` For other configuration parameters such as multicast addresses, ports, and NMOS mode, see the [Configuration Reference](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/latest/configuration.html). ## Step 5. Verify Deployment and View Output Check pod status: ```bash kubectl get pods ``` Confirm sender, NIM, and receiver pods are `Running` and `READY 1/1`. To preview the enhanced audio output over SRT: ```bash kubectl get nodes -o wide ``` Identify the node hosting the receiver pod with `kubectl get pods -o wide`, then read that node's `INTERNAL-IP` from `kubectl get nodes -o wide`. Open **VLC** → **Media** → **Open Network Stream**, then paste `srt://:32090`. `32090` is the default SRT NodePort exposed by the Helm chart and can be configured during installation. Or use ffplay: ```bash ffplay "srt://:32090" ``` ## Step 6. Uninstall and Cleanup Remove the release: ```bash helm uninstall studio-voice-h4m-sample ``` For additional deployment options (NMOS mode, operator, standalone NIM service), and advanced configuration, refer to the [documentation](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/latest/index.html). 39:[["$","div",".3",{"className":"relative","children":[["$","div",null,{"className":"relative flex min-h-[194px] w-full flex-col overflow-hidden p-md sm:p-ml bg-manitoulinLightBlack","children":[["$","$L3c",null,{"alt":"","className":"object-cover object-center !left-1/2 z-10 !w-1/2 brightness-20 lg:brightness-75","fill":true,"priority":true,"src":"https://assets.ngc.nvidia.com/products/api-catalog/images/studiovoice.jpg","style":{"maskImage":"linear-gradient(to left, rgba(0,0,0,1), rgba(0,0,0,0))"}}],["$","div",null,{"className":"z-30 mx-auto flex w-full max-w-screen-xl min-w-0 flex-1 flex-col items-start justify-between gap-md lg:flex-row lg:gap-ml","children":["$","section",null,{"aria-describedby":"resource-description","aria-labelledby":"resource-name","className":"flex w-full min-w-0 flex-col lg:flex-1","children":["$","div",null,{"className":"flex flex-col gap-1.5 whitespace-pre-wrap","children":[["$","$L3d",null,{"className":"w-fit text-ml text-n300 lowercase hover:text-n050","href":"/nvidia","children":"NVIDIA"}],["$","div",null,{"className":"flex flex-col items-start gap-3 lg:flex-row lg:items-center","children":[["$","h1",null,{"className":"max-w-full text-lg leading-heading font-medium break-words text-primary","data-testid":"resource-name","id":"resource-name","children":"Studio Voice"}],[["$","span","AVAILABLE",{"className":"nv-badge nv-badge--kind-solid h-[18px] py-0","data-testid":"nv-badge","ref":"$undefined","children":["$undefined","Downloadable"]}],["$","span","PREVIEW",{"className":"nv-badge nv-badge--kind-solid nv-badge--color-purple h-[18px] py-0","data-testid":"nv-badge","ref":"$undefined","children":["$undefined","Free Endpoint"]}]],"$undefined"]}],["$","p",null,{"className":"max-w-[85ch] text-ms text-primary","id":"resource-description","children":"Enhance input speech recorded with low-quality microphones in noisy or reverberant environments, producing studio-quality speech."}],["$","div",null,{"className":"flex w-full flex-col items-start gap-md lg:flex-row lg:items-end lg:justify-between","children":[["$","div",null,{"className":"flex w-full min-w-0 flex-wrap gap-sm lg:flex-1","children":["$","$L3e",null,{"artifactType":"ENDPOINT","density":"standard","labels":["2110","broadcast","communications","mic quality","nimType:endpoint:nim_type_preview","nimType:endpoint:nim_type_run_anywhere","nvidia ai for media","nvidia holoscan for media","smpte","speech enhancement"],"searchRoutePath":"/models"}]}],["$","div",null,{"className":"z-30 flex w-full flex-row justify-start gap-ms sm:justify-end lg:w-auto lg:shrink-0","children":["$undefined",["$","$L3f",null,{"cta":{"text":"Apply to Self-Host","url":"https://www.nvidia.com/en-us/ai/nim-notifyme/","nim_available_override_url":"https://catalog.ngc.nvidia.com/orgs/nim/teams/nvidia/containers/maxine-studio-voice"},"isNIMAccessRestricted":"$undefined","isNIMAvailable":true}]]}]]}]]}]}]}],["$","div",null,{"className":"pointer-events-none absolute right-0 bottom-0 left-0 z-20 h-30 bg-gradient-to-t to-transparent from-manitoulinLightBlack via-manitoulinLightBlack/70"}]]}],["$","div",null,{"className":"px-4 mx-auto flex max-w-screen-xl flex-col pt-0 lg:px-0","data-body-background":"light-black","children":["$","section",null,{"className":"flex w-full flex-col justify-between rounded-4 border-manitoulinBorderColor bg-manitoulinDarkBlack sm:rounded-6 lg:rounded-8 p-ms lg:p-lg","children":["$","$L40",null,{"artifact":{"artifactType":"ENDPOINT","name":"studiovoice","displayName":"Studio Voice","publisher":"nvidia","shortDescription":"Enhance input speech recorded with low-quality microphones in noisy or reverberant environments, producing studio-quality speech.","logo":"https://assets.ngc.nvidia.com/products/api-catalog/images/studiovoice.jpg","labels":"$39:0:props:children:0:props:children:1:props:children:props:children:props:children:3:props:children:0:props:children:props:labels","attributes":[{"key":"AVAILABLE","value":"true"},{"key":"PREVIEW","value":"true"},{"key":"last_month_api_invocation_count","value":"5644"},{"key":"last_month_download_count","value":"0"}],"updatedDate":"2025-06-13T14:12:57.045Z","bias":"| Field | Response |\n|-------|----------|\n| Participation considerations from adversely impacted groups [protected classes](https://calcivilrights.ca.gov/disputeresolution/protected-characteristics/) in model design and testing: | Age (18+), Gender |\n| Measures taken to mitigate against unwanted bias: | Evaluated using internal, proprietary data mix to achieve similar key performance indicators. |","canGuestDownload":true,"createdDate":"2024-10-03T00:40:13.315Z","description":"$41","explainability":"$42","isPublic":true,"isReadOnly":true,"orgName":"qc69jvmznzxy","privacy":"$43","safetyAndSecurity":"| Field | Response |\n|---------|------------|\n| Model Application(s): | Speech Enhancement |\n| Describe the life critical impact (if present): | Not Applicable |\n| Use Case Restrictions: | Abide by [NVIDIA Open Model License](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/). |\n| Model and dataset restrictions: | The Principle of Least Privilege (PoLP) is applied limiting access for dataset generation and model development. Restrictions enforce dataset access during training, and dataset license constraints adhered to. |"},"endpointName":"studiovoice","endpointSpec":{"namespace":"qc69jvmznzxy","updatedDate":"2026-08-10T20:30:04.158Z","nvcfFunctionId":"3f0aeba3-6d91-4465-b8cc-cc2aef355186","createdDate":"2024-10-03T00:40:13.518Z","attributes":{"apiDocsUrl":"https://docs.nvidia.com/nim/maxine/studio-voice/latest/index.html","termsOfUse":"GOVERNING TERMS: This trial service is governed by the NVIDIA API Trial Service Terms of Use. Use of this model is governed by the NVIDIA Open Model License.\n","showUnavailableBanner":false,"cta":"$39:0:props:children:0:props:children:1:props:children:props:children:props:children:3:props:children:1:props:children:1:props:cta","usage":"$44","deploy":[{"label":"Linux with Docker","filename":"linux.md","contents":"$45"},{"label":"Deploy on Holoscan for Media Platform","filename":"h4m.md","contents":"$46"}]},"artifactName":"studiovoice"},"isLoginRequired":true,"playgroundElement":"$47","publisherName":"nvidia","searchParams":{"environment":"wsl2.md","model":"studiovoice"},"tabs":[{"label":"Experience","bare":true,"icon":"play","kind":"default","id":"experience","href":"/nvidia/studiovoice","aliases":["/nvidia/studiovoice/experience"]},{"label":"Model Card","icon":"book","kind":"default","id":"modelcard","href":"/nvidia/studiovoice/modelcard","aliases":"$undefined"},{"label":"Try API","kind":"default","id":"api","href":"/nvidia/studiovoice/api","aliases":"$undefined"},{"label":"Deploy","kind":"default","id":"deploy","href":"/nvidia/studiovoice/deploy","aliases":"$undefined"}]}]}]}]]}]] 3a:["$","$L48",null,{"spec":"$39:0:props:children:1:props:children:props:children:props:endpointSpec"}]