AI Audio Workflows in Rivya: From Script to Delivery
Create speech, dialogue, custom music, sound clips or lyric text, and transcribe a recording. Check each model’s inputs and credit estimate.
Last reviewed on November 11, 2026
Create narration or spoken dialogue in Audio Studio, generate sounds with Suno Sounds, or draft text with Suno Lyrics. Suno Music supports new custom-mode generation. Music extension, WAV conversion and vocal separation from Music results remain unavailable. You can still revisit existing Music tasks, check their status and open their media while the files remain accessible. Lyrics creates text and does not start a Music task.
For Suno Music, enter a title and music style. For vocals, enter the words to sing in the lyrics field; for an instrumental, leave lyrics empty. Choose the model version and target duration offered by the form. This does not guarantee exact timing, lyrics or arrangement. Review the model page and credit estimate, then submit once.
For the form fields, submission steps and result controls, read the Audio Studio guide. For a first-task walkthrough, read starting an AI audio workflow.
Suno Speech: scenes and supplied scripts
Suno Speech — Suno Speech creates spoken audio with tone, rhythm and emotion. Simple mode writes a script from your scene; Advanced mode reads the script you provide.
Choose Simple to describe an announcement, introduction or spoken scene and let the model write it. Choose Advanced when the wording matters: your main text becomes the script, with optional delivery settings. Switching to Simple drops the Advanced settings.
Suno Speech generates expressive spoken audio and can add background music. For a script with one of the named Zephyr, Puck or Kore voices and WAV output, choose an available Gemini TTS model instead.
Gemini 3.8 Flash TTS · Gemini 3.8 Flash Lite TTS · Gemini 3.1 Flash TTS · Gemini 2.5 Pro TTS
Short-script Gemini speech
The four Gemini TTS models are available for a single speaker reading one short script: 3.8 Flash-Lite, 3.8 Flash, 3.1 Flash and 2.5 Pro. Choose Zephyr, Puck or Kore. The original script is limited to 1,024 UTF-8 bytes, including spaces and line breaks, rather than 1,024 characters. A successful task returns WAV audio.
This path does not accept multiple speakers, attachments, voice cloning or additional scene controls. The reserves are 15, 22, 48 and 48 credits respectively, not promised final charges. Confirmed lower consumption returns the difference; unresolved consumption stays under review. If submission is uncertain, check the existing task instead of submitting it again.
Start with the material you already have
| Starting material | What you want | Route |
|---|---|---|
| A finished script | One person reading it | ElevenLabs Turbo 2.5 or ElevenLabs Multilingual V2; check the model's supported languages and voices |
| A script with speaker turns | A spoken conversation | ElevenLabs Dialogue V3; assign text and a voice to every turn |
| A description of music | A song or instrumental draft | Suno Music — Custom generation available |
| A description of a sound | Ambience, a loop or a short sound sketch | Suno Sounds |
| A topic for a song | Lyrics to review before making music | Suno Lyrics; the result is text |
| An existing recording | Cleaner speech with the original performance preserved | No currently available direct cleanup route; Audio Isolation is paused |
| An existing recording | A plain-text transcript | Nemotron 3.5 Transcribe; upload one MP3 or WAV. No word timestamps, translation or video input |
ElevenLabs Audio Isolation and Sound Effect V2 are currently unavailable. Check model status before choosing a route. Suno Sounds generates new audio from a description; it does not clean an uploaded recording or offer every control of a dedicated sound-effect model.
Finish spoken words before choosing delivery settings
For narration, write the words the listener should hear. Keep creative instructions separate from that text: a request such as “make a calm introduction” may itself be read aloud. For dialogue, split the script into turns and choose a voice for each one.
For another language, preserve the meaning while adapting wording, sentence length and idioms for listening. Check names, numbers and abbreviations aloud. Changing the website language does not translate your script, and the model's supported languages may differ from Rivya's interface languages. The narration guide covers spoken explanations; the dubbing guide covers replacing dialogue in another language.
Keep lyrics, music and sound generation separate
Describe the sound you need for Suno Sounds. Use the available loop, tempo and musical-key settings when they matter to the scene. A prompt or setting does not guarantee an exact duration or arrangement; listen to the result before using it.
Use Suno Lyrics for a lyric brief, then review and edit the returned text. Lyrics does not produce a sung track or start a Music task.
Music descriptions and instrumental settings apply to Suno Music. Suno Music supports new custom-mode generation. Music extension, WAV conversion and vocal separation from Music results remain unavailable. Check the model's current status and form before planning a new Music task.
Eligible successful Suno Sounds results can offer vocal separation when the required track information is present. The action starts a new task with a separate credit charge. It cannot take an arbitrary uploaded recording, and separation from Music results is paused. See music workflows.
Example: prepare narration and music for a product video
For a 20-second product video, you might use the following sequence. It is a planning example; Rivya does not assemble the video for you:
Finish the scene order and spoken script. Decide which moments need speech, music or silence; keep pronunciation notes separately.
Generate a spoken draft from the script. Listen against the actual scenes and adjust wording or available speed settings where needed. Asking for a duration in the script does not enforce exact timing.
Create a new Suno Music draft in custom mode or use a Sounds result. Listen to the full file and its ending, save a copy, and combine separate audio and video in your editor.
Download the accepted files, align and mix them in your editor, then review the complete video. Rivya does not automatically combine these separately generated tracks or synchronize translated speech with the original video.
For a recording that will drive a speaking character or lip-sync an existing video, start with audio uploads. Check the selected model's companion image or video requirements and upload limits. These routes produce video rather than narration audio. See voiceover for videos for the editorial workflow.
Budget for the whole sequence
Speech estimates depend on the main text length; dialogue estimates use the total text across turns. Sounds, Lyrics and eligible Sounds vocal separation have separate task prices. Review the estimate before submitting and the task's reserved or final credits afterward. Suno Music supports new custom-mode generation. Music extension, WAV conversion and vocal separation from Music results remain unavailable. See credit rules.
Use a representative short draft to check a voice before committing to a longer script. A draft still costs credits, and it does not guarantee identical delivery in a later result. If a submission response is unclear, check the original task in History before submitting again. Follow failed-task and credit-return guidance for a recorded failure; disliking an otherwise completed result is a different case.
Keep the files needed for the next stage
Save the accepted audio, final script, language version and any pronunciation or editing notes together. Preserve original recordings separately. Before delivery, listen for missing words, wrong names, unexpected vocals and timing problems; confirm you can use the source material and result for the intended purpose. A paid plan does not grant rights to someone else's voice or protected work.
Save useful results promptly while their media links are accessible; see output downloads. History saves task records; keep your own audio files and editing project, because the media links may stop working later. Keep an external copy of unsent text and settings when moving between a model page and Studio; see public pages and signed-in workflows.
