Text to speech
Bring: The exact words to be spoken.
Review: An audio reading of the supplied text.
Text to speech · Model and workflow guide
An accurate spoken reading of supplied words.
Compare text to speech options for an accurate spoken reading of supplied words. Review source requirements, candidate details and current Promptha support.
Compare model options →Start with the source and the result.
A concrete brief
Welcome to Harbor Roast. Discover our autumn blend at the neighborhood market.
Illustrative workflow brief. It is not a tested model result or a universal API input; supply each model’s required source fields separately.
What to check in a candidate
- Word accuracy
- Names and dates
- Pauses
Model options for this work
Ordered by proposed catalog placement, not benchmark performance. Check each entry’s integration status.
MiniMax Speech 2.8 HD
Speech and expressive voiceovers
34 credits · ≤ $0.10 provider estimate
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · Integration preview
- Chat → Voiceover · Integration preview
Pricing by action
Conservative USD provider estimates and Promptha credits for the stated examples. Exact inputs and settings receive a fresh quote before submission.
- Voiceover · 34 credits · ≤ $0.10 provider estimate · 1,000 UTF-8 script bytes at default voice settings. Policy snapshot 2026-10-03. Exact request is quoted before submission. Provider pricing ↗
Explore linked prompts, profiles and templates →
Suggested uses
- Advertising narration
- Character voices
- Multilingual speech
- Storytelling
- Catalog role
- Hero candidate · proposed premium role
- Provider verification
- Input schema checked 2026-10-03
- Promptha support
- speech.generate
Adapter and price policy are implemented; workspace activation and live output verification are separate. Provider features may exceed the controls exposed in Promptha.
Registered action adapters. Check workspace availability; Preview results still need live output verification.
Provider input fields
Provider schema fields are not a list of enabled Promptha controls.
text, pitch, speed, volume, bitrate, channel, emotion, voice_id, sample_rate, audio_format, language_boost, subtitle_enable, english_normalization
Required by schema: text
Try the connected workflow
Prepare a request in Chat, add any required source and review it. This does not start generation.
Prepare Voiceover in Chat →Google Gemini 3.1 Flash TTS
Explore Google Gemini 3.1 Flash TTS for quick narration, assistant speech, voiceovers. Review its inputs and Promptha integration status.
Pricing not verified
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · planned for this model
- Chat → Voiceover · planned for this model
Explore linked prompts, profiles and templates →
Suggested uses
- Quick narration
- Assistant speech
- Voiceovers
- Catalog role
- Recommended candidate · proposed default role
- Provider verification
- Input schema checked 2026-09-24
- Promptha support
- No execution adapter yet
Catalog candidate only. This model cannot be selected for generation through this catalog.
Pricing not verified. This listing cannot authorize generation.
Provider input fields
Provider schema fields are not a list of enabled Promptha controls.
text, voice, prompt, language_code
Required by schema: text
Inworld Realtime TTS 2
Speech and expressive voiceovers
9 credits · ≤ $0.025 provider estimate
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · Integration preview
- Chat → Voiceover · Integration preview
Pricing by action
Conservative USD provider estimates and Promptha credits for the stated examples. Exact inputs and settings receive a fresh quote before submission.
- Voiceover · 9 credits · ≤ $0.025 provider estimate · 1,000 UTF-8 script bytes at default voice settings. Policy snapshot 2026-10-03. Exact request is quoted before submission. Provider pricing ↗
Explore linked prompts, profiles and templates →
Suggested uses
- Game dialogue
- Expressive characters
- Interactive speech
- Catalog role
- Recommended candidate · proposed specialist role
- Provider verification
- Input schema checked 2026-10-03
- Promptha support
- speech.generate
Adapter and price policy are implemented; workspace activation and live output verification are separate. Provider features may exceed the controls exposed in Promptha.
Registered action adapters. Check workspace availability; Preview results still need live output verification.
Provider input fields
Provider schema fields are not a list of enabled Promptha controls.
text, language, voice_id, sample_rate, temperature, audio_format, speaking_rate, text_normalization
Required by schema: text
Try the connected workflow
Prepare a request in Chat, add any required source and review it. This does not start generation.
Prepare Voiceover in Chat →MiniMax Speech 2.8 Turbo
Speech and expressive voiceovers
20 credits · ≤ $0.06 provider estimate
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · Integration preview
- Chat → Voiceover · Integration preview
Pricing by action
Conservative USD provider estimates and Promptha credits for the stated examples. Exact inputs and settings receive a fresh quote before submission.
- Voiceover · 20 credits · ≤ $0.06 provider estimate · 1,000 UTF-8 script bytes at default voice settings. Policy snapshot 2026-10-03. Exact request is quoted before submission. Provider pricing ↗
Explore linked prompts, profiles and templates →
Suggested uses
- Speech prototypes
- Social voiceovers
- Narration drafts
- Catalog role
- Recommended candidate · proposed fast role
- Provider verification
- Input schema checked 2026-10-03
- Promptha support
- speech.generate
Adapter and price policy are implemented; workspace activation and live output verification are separate. Provider features may exceed the controls exposed in Promptha.
Registered action adapters. Check workspace availability; Preview results still need live output verification.
Provider input fields
Provider schema fields are not a list of enabled Promptha controls.
text, pitch, speed, volume, bitrate, channel, emotion, voice_id, sample_rate, audio_format, language_boost, subtitle_enable, english_normalization
Required by schema: text
Try the connected workflow
Prepare a request in Chat, add any required source and review it. This does not start generation.
Prepare Voiceover in Chat →Chatterbox Turbo
Explore Chatterbox Turbo for voice experiments, character speech, narration concepts. Review its inputs and Promptha integration status.
Pricing not verified
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · planned for this model
- Chat → Voiceover · planned for this model
Explore linked prompts, profiles and templates →
Suggested uses
- Voice experiments
- Character speech
- Narration concepts
- Catalog role
- Specialist · proposed experimental role
- Provider verification
- Input schema checked 2026-09-24
- Promptha support
- No execution adapter yet
Catalog candidate only. This model cannot be selected for generation through this catalog.
Pricing not verified. This listing cannot authorize generation.
Provider input fields
Provider schema fields are not a list of enabled Promptha controls.
seed, text, top_k, top_p, voice, temperature, reference_audio, repetition_penalty
Required by schema: text
ElevenLabs V3
Speech and expressive voiceovers
34 credits · ≤ $0.10 provider estimate
Uses, inputs and support
Where it fits in Promptha
- Chat → Voiceover, or document-to-speech action · Integration preview
- Chat → Voiceover · Integration preview
Pricing by action
Conservative USD provider estimates and Promptha credits for the stated examples. Exact inputs and settings receive a fresh quote before submission.
- Voiceover · 34 credits · ≤ $0.10 provider estimate · 1,000 UTF-8 script bytes at default voice settings. Policy snapshot 2026-10-03. Exact request is quoted before submission. Provider pricing ↗
Explore linked prompts, profiles and templates →
Suggested uses
- Speech and expressive voiceovers
- Catalog role
- Specialist · proposed specialist role
- Provider verification
- Input schema checked 2026-10-03
- Promptha support
- speech.generate
Adapter and price policy are implemented; workspace activation and live output verification are separate. Provider features may exceed the controls exposed in Promptha.
Registered action adapters. Check workspace availability; Preview results still need live output verification.
Try the connected workflow
Prepare a request in Chat, add any required source and review it. This does not start generation.
Prepare Voiceover in Chat →Generator ideas
Proposed recipes to configure and validate before offering them as runnable workflows.
Voiceover Generator
A recording of the approved script. Supply the source material, review a model-specific request and inspect the resulting candidate before delivery.
Suggested stages: Voice generation → Text to speech.
AI Voiceover
A voice recording from supplied text. Supply the source material, review a model-specific request and inspect the resulting candidate before delivery.
Suggested stages: Voice generation → Text to speech.
Choose a model, then review the work.
Use the model guide to check supported inputs. Keep the source available, compare the complete candidate and continue from the exact version you choose. Credit pricing and live output verification are required before activating new integrations.
THE USEFUL DETAILS
Good to know.
Creation notes
Prepare text to speech inputs
The starting material for this capability is: The exact words to be spoken. The intended result is: An audio reading of the supplied text. Check the selected model's exposed input contract before preparation. A visually related source may still be incompatible in media type, size or the role it plays in the request.
Review these concrete details: word accuracy, names and dates, pauses. Keep the approved brief and original source available. Inspect the complete candidate rather than judging a thumbnail, single frame or short excerpt. Record the exact version you choose and keep rejected alternatives when they explain the decision.
Read ElevenLabs Multilingual v2 integration guideChoose a connected route or keep the work in research
Some related operations have registered integration guides. Inspect each action and workspace activation; registration does not guarantee that all provider controls or the complete capability workflow are available.
A proposed Generator recipe still needs saved configuration, compatible stages and reviewed inputs. Inspect the quote and the real output handling before treating a multi-step process as a supported deliverable. Export and native Studio compatibility belong to the chosen artifact, not the name of the capability.
Inspect catalog support for this capabilityCommon questions
Can I run text to speech in Promptha?
Some related operations have registered adapters. Check the specific model guide, required sources and live workspace choices. The capability label does not guarantee every proposed workflow is enabled.
What should I inspect in the result?
Text to speech: Word accuracy, Names and dates, Pauses. Compare the complete candidate with its approved brief and source before choosing it.