About
Boston Children's Hospital ALS Program's Messagebank to Voicebank initiative — powered by Acapela Group's voice banking platform — enables individuals with ALS and other degenerative speech conditions to 'double dip' by leveraging the same recordings made for message banking to simultaneously build a personalized synthetic voice bank. Message banking captures meaningful, emotionally significant phrases in a person's real voice for later playback, while voice banking uses recorded speech samples to train an AI-synthesized voice model that can generate new utterances on demand. By combining both workflows, users maximize the value of every recording session without doubling the effort. The Acapela voice banking engine analyzes the recorded samples to construct a custom text-to-speech voice that closely mimics the speaker's natural timbre, rhythm, and accent. The resulting voice can be integrated into AAC (Augmentative and Alternative Communication) devices and software, allowing individuals who have lost the ability to speak to continue communicating in their own recognizable voice. This approach is particularly valuable for ALS patients, who often face a narrow and shrinking time window in which their speech is still sufficiently clear for quality recordings. The dual-use methodology reduces recording burden while ensuring both immediate (playback phrases) and long-term (generative TTS) communication needs are addressed, supporting dignity and personal identity through voice preservation.
Key Features
- Double-Dip Recording Workflow: Recordings made for message banking are simultaneously used to train a personalized synthetic voice, eliminating the need for two separate recording sessions.
- Custom AI Voice Synthesis: Acapela's engine analyzes speech samples to generate a text-to-speech voice that replicates the speaker's unique timbre, accent, and speech patterns.
- AAC Device Integration: The generated voice bank is compatible with AAC (Augmentative and Alternative Communication) devices and software, enabling ongoing communication after voice loss.
- Message Playback Library: Meaningful phrases and personal messages are stored for immediate, emotionally authentic playback — preserving the speaker's real voice for key moments.
Pros
- Reduces Recording Burden: The dual-use approach means ALS patients need fewer total recording hours to achieve both message banking and voice banking goals.
- Preserves Personal Identity: Users communicate in their own synthesized voice rather than a generic one, maintaining a sense of self and recognition by family and caregivers.
- Clinically Supported: Developed in partnership with Boston Children's Hospital ALS Program, providing medically informed guidance on timing and methodology.
Cons
- Narrow Recording Window: Voice quality depends on recording early enough in the ALS progression; delayed enrollment can result in a voice bank that sounds noticeably degraded.
- Limited Platform Access: The service is primarily web-based with no dedicated mobile app, which may be restrictive for users with advanced motor impairments.
- Output Quality Varies: The fidelity of the synthesized voice depends heavily on the quantity and quality of recordings provided, requiring sufficient effort during the recording phase.
